Looking for adverse reactions

The monitoring needed for adverse reactions to drugs used in HIV treatment depends on which drugs are being used and how they are used. It may also vary with nutritional status and between populations exposed to different infections.

The main reason for monitoring adverse reactions is to ensure the safety of individuals taking treatments and identify those who need to stop treatments that are damaging their health and change to other treatments, if available.

The drug safety regulatory bodies in many countries operate reporting schemes for doctors and pharmacists to report adverse events which they suspect may be linked to treatments. This is known as pharmacovigilance and is important for identifying what may be rare but serious side effects of drugs that are not picked up in clinical trials.

When drugs are taken in combination, deciding what contribution each of those drugs has to effects that are seen, as well as trying to decide what effects are due to illness, or are completely unrelated either to the drugs or to the illness, is a challenge.

An example of this is the discussion that is still taking place as to what causes lipodystrophy the changes in body-fat distribution that are seen among people with HIV who are on long-term treatment.

NAMs Information series for people HIV includes a booklet on lipodystrophy, and is available in pdf format on aidsmap.com.

Baseline assessments and screening tests

To monitor drug reactions, it is essential that baseline assessments and measurements are made, before the drug is first taken. Good medical record keeping is fundamental for effective monitoring.

Some people may be more vulnerable to particular side effects from drugs than others. Some of these factors can be identified through a careful clinical history. Being overweight, being malnourished, or drinking alcohol can all predispose some people to particular drug side effects. A past history of illness such as pancreatitis is a major risk factor for its occurrence as a drug side effect.

People who have some peripheral neuropathy before starting on treatment are specially vulnerable to it. A careful physical examination for lack of sensation is therefore recommended before starting on drugs such as isoniazid (INH), d4T (stavudine), ddI (didanosine) or ddC (zalcitabine) in particular.

Other conditions, such as hepatitis B and C, which increases the risk of liver damage with drug treatment, require blood tests for accurate diagnosis. In the absence of blood tests, it may still be possible to assess likely risk for example, injecting drug users and people who have received blood products are more likely to be hepatitis C positive than people whose exposure to HIV was through sex. Some populations, notably in China, have a very high prevalence of hepatitis B virus acquired at birth.

In Cote d'Ivoire, a strong correlation emerged between hepatitis B virus co-infection and hepatotoxicity as measured every three months using serum alanine aminotransferase (ALAT) tests, in a population of people with HIV receiving ARVs including protease inhibitors. There was no similar relationship for hepatitis C virus in this population, although the small numbers (9 with HCV and 21 with HBV out of 112 patients with elevated ALAT values) do not allow definitive conclusions (Dakoury).

It may still be necessary to offer a treatment despite a known risk of side effects, but it is then possible for both the patient and healthcare provider to be aware of the risk and to discuss how it can be managed.

Clinical monitoring vs laboratory tests

Much of the monitoring that is needed can be carried out through clinical examinations and history-taking, and through telling patients what to expect, what to do if they have problems, and when they need to report problems to a doctor or nurse or other health care worker.

It has been claimed that an experienced doctor can identify virtually all known drug toxicities for ARVs without access to laboratory tests. The problem is, that a radical expansion of access to ARVs must depend on less experienced doctors and other healthcare workers. Even so, when one or two standard first-line treatments are used, it is possible to train health care workers to recognise the most common side effects and the most serious adverse events linked to the drugs used.

What is usually understood by monitoring does require some access to laboratory tests, most of which require a blood sample to be taken.

Are these really necessary, and how much difference do they make to the outcome for patients? This question may be addressed directly, through clinical trials comparing outcomes for groups of patients with and without access to laboratory monitoring tests.

A preliminary trial of this kind has been reported from Uganda (Enzama), where 100 ARV naive patients (37 male, average age 39) with initial CD4 count <200 were randomised to receive monthly clinical monitoring only - CMO (50 pts) and clinical monitoring plus laboratory monitoring at three monthly intervals (50 pts). Outcomes in the two arms were compared. Clinical monitoring consisted of clinical history, body weight, temperature, [blood?] pressure and checking for symptoms of opportunistic infections. Laboratory monitoring consisted of viral load, CD4/CD8, renal and liver function tests for toxicity. These tests required use ofa PCR machine, flow cytometer, and automated chemistry analyser respectively, plus highly trained personnel. Blinded CD4/CD8 counts were performed on those given 'clinical monitoring only' every 3 months, for comparison. 5 patients moved from grade 2 adverse events to grade 3 in the group given clinical monitoring only, and 6 in the other arm. There were no deaths in either arm. There were no significant differences in prognosis in both arms.

Laboratory tests could be divided into two categories.

• Routine tests that should be carried out, whether or not there are symptoms.

• Tests that should be available when symptoms are reported by a patient and particular problems are suspected.

Which of the tests listed in the first section, above, should be considered routine, and which used only when problems are suspected, may vary depending on the kinds of treatments used, the populations seen, the clinical experience of the team, and other factors.

For example, liver function will be more important, the higher the prevalence of liver disease such as hepatitis C. It will be important to monitor it for at least a couple of months when nevirapine or efavirenz are being used for long-term treatment. (See Liver function in A to Z of medical tests for more details.)

Similarly, anaemia (red blood cell counts) will be most important when AZT is in use and in populations particularly vulnerable to anaemia, such as pregnant women in countries where malaria and other parasitic diseases are common. (See Blood count in A to Z of medical tests for more details.)

A sample protocol from Morocco

A Moroccan treatment centre that has provided ARV treatment, mostly including protease inhibitors with nucleoside analogues, adopted the following protocol for monitoring for toxicity.

All patients started on antiretrovirals had blood taken before starting and then at two weeks, 1 month, 2 months, 6 months and 9 months.

Tests undertaken: haemoglobin; ALT (a liver function test); renal function; creatine kinase; amylase; blood lipids.

Predictable short term side effects

Some side effects are common, generally short-term and should not normally lead to treatment being stopped. It is important that patients are warned that they may experience these (although many people will not), and are encouraged to treat themselves appropriately or seek help rather than stop taking the drugs.

Such side effects include mild forms of nevirapine-related skin rash, initial headache and nausea with AZT and a range of gastrointestinal disturbances with protease inhibitors and other ARVs. Bad dreams and spaced out sensations with efavirenz may also fall into this category.

Serious and longer-term side effects

Other side effects that can occur in the short term or develop over time are serious and should lead to review of treatment. These may include severe skin rashes, pancreatitis, liver damage and other conditions linked to particular drugs.

In the case of abacavir, there is a hypersensitivity reaction which seems to affect up to two per cent of people who take it, sometimes with a rash, fever, sometimes feeling flu-like. If this is suspected, it is very important that the drug is stopped and that the persons medical notes are clearly marked, that they should never take the drug again.

Click here for further information about abacavir hypersensitivity reaction.

The severe liver toxicity that is occasionally seen with nevirapine, in the first few weeks of treatment, appears to be due to an immune reaction, and to be more common among people with HIV who start the drug at a relatively high CD4 count. Arguably, it has more in common with the abacavir hypersensitivity reaction than it does with the liver toxicity associated with other ARVs such as protease inhibitors.

Skin rashes, which can be serious and even life-threatening in the case of nevirapine, delavirdine, cotrimoxazole, do not require laboratory tests.

Many other side effects, however, may do so.

Blood counts

There is a known risk of anaemia from AZT (zidovudine) which may be greater when treatment is given in advanced disease and with a low CD4 count, than earlier in the disease. Anaemia can also occur as a result of HIV disease itself. The protease inhibitor indinavir has been linked to a small number of cases of anaemia.

White blood cell counts can also be damaged by AZT as part of a general suppression of bone marrow activity.

Liver function tests

Routine blood tests for liver function are necessary when nevirapine is used as part of ongoing therapy and for protease inhibitors, especially where coinfection with hepatitis C or hepatitis B is or may be present. Other medicines can cause liver damage include cotrimoxazole, rifampicin and rifabutin. Liver damage can also happen from using alcohol or recreational drugs.

Serum albumin tests look for a protein that is made by the liver and released into the blood to keep blood circulating rather than being drawn into the tissues. Reduced levels of albumin suggest that the liver is not working as it should.

Low serum albumin is also a marker for poor nutritional status, and in some populations it may predict progression in HIV disease. This may be a reason for using the test, to identify people at higher risk who should have priority access to treatment and other medical and social support. On the other hand, it may reduce its value as a marker for liver damage (Feldman).

Bilirubin is a protein released into the blood by red blood cells when they are destroyed at the end of their lives. Bilirubin is the cause of jaundice, when a persons eyeballs and skin go yellow and their urine is darkened. Jaundice happens when the liver is no longer doing its job of removing the bilirubin from the blood. Increased levels of bilirubin can be detected before jaundice is obvious, suggesting that liver damage is occurring. Bilirubin tests are recommended when indinavir is being taken.

Liver enzyme tests are advised when treatment includes:

• d4T (stavudine),

• nevirapine

• delavirdine

• any protease inhibitor.

These tests detect proteins that are made by the liver and normally are found only or mainly in the liver. When the liver is damaged, these enzymes are released into the blood. The higher the levels, the greater the damage to the liver. The two most useful tests are for alanine aminotransferase (ALT or GPT) and aspartate aminotransferase (AST or GOT). A similar enzyme, which is less liver-specific, is alkaline phosphatase (ALK PHOS).

Metabolic complications

Cholesterol levels in the blood are raised with long-term treatment with a number of the protease inhibitors currently in use and possibly with other drugs. Whether this will translate into a substantial risk of heart disease probably depends on the extent to which other risk factors (such as cigarette smoking, family history, etc) are present. Triglycerides are another kind of lipid in the blood, which increase the risk of pancreatitis, as discussed below (and see Lipid test in A to Z of medical tests for more details).

Urine tests for glucose can be used to monitor insulin resistance a low-level form of diabetes which has been reported with protease inhibitor treatment.

Lactic acidosis is a rare and very serious complication linked to nucleoside analogues in general although it may be more common with AZT (zidovudine), d4T (stavudine) and ddI (didanosine) than with abacavir or 3TC (lamivudine). It is caused by damage to mitochondria the systems inside cells that generate energy by combining sugars and oxygen. This makes the body switch to alternative energy systems that cause lactic acid to build up in the blood. The main symptoms are fatigue, breathlessness, and liver enlargement. Initial symptoms can include nausea, lack of appetite and malaise, as well as fatigue and difficulty in breathing. Muscle pain and numbness or tingling sensations have also been reported. In lactic acidosis, the liver may be swollen and tender, and liver enzymes may be elevated. Symptoms of acute lactic acidosis include difficulty breathing and hyperventilation.

See Lactic acidosis / acidaemia in Symptoms and illnesses: A to Z of illnesses for further information on this problem.

Pancreatitis is a rare but very serious complication mainly linked to ddI (didanosine) and to a much lesser extent d4T (stavudine) and ddC (zalcitabine). It is possible to monitor early signs by checking levels of the enzyme amylase in the blood, before symptoms develop. The usual symptoms include nausea, vomiting and sometimes pain (which can be very severe) in the region around the stomach just below where the ribs join in the centre of the chest. The abdomen may be very tender and painful to the touch. If pancreatitis is suspected, urgent medical help must be sought.

See Pancreatitis in the A to Z of illnesses for further details.

Triglycerides in the blood are monitored in many treatment centres that use antiretrovirals, primarily because raised levels which can occur with protease inhibitors and other ARVs greatly increase the risk of pancreatitis.

Kidney damage

The protease inhibitor indinavir causes kidney problems which can be prevented to some degree by maintaining high leves of fluid intake through the day. Combining ritonavir with indinavir removes some of the problems associated with taking indinavir (especially dietary restrictions) but does not remove the risk of kidney damage. Ritonavir itself may cause kidney damage, which should be monitored. Renal function tests are carried out to check that the kidneys are working properly.

Two tests commonly performed to monitor how the kidneys are working are creatinine levels and BUN (blood urea nitrogen) tests. If these are elevated, it suggests the kidneys may be damaged.

People who are taking indinavir need to be informed of the need to drink extra water and encouraged to report any symptoms suggestive of kidney problems promptly.

If kidney stones are suspected, then a urinary test for erythrocytes (red blood cells) will show haematurea if it isnt obvious.

References

Dakoury C et al. Role of viral infections HCV and HBV in liver toxicity among patients receiving antiretroviral treatment in HIV drugs access initiative in Cote d'Ivoire. XIV International AIDS Conference, Barcelona, abstract MoPeB3280, 2002.

Enzama R et al. Alternative methods for therapeutics monitoring. Second International AIDS Society Conference on HIV Pathogenesis and Treatment, Paris, abstract 694, 2003.

Feldman JG et al. Serum albumin as a predictor of survival in HIV-infected women in the Women's Interagency HIV Study. AIDS 14: 863-870, 2000.