Management of hyperlactatemia: no need for routine lactate measurements

Despite the impressively beneficial effects of antiretroviral medication for HIV-infected patients, it has become increasingly evident that prolonged therapy can also have adverse effects. One of the most fearful is the development of severe lactic acidosis, induced by the mitochondrial toxicity of nucleoside analogue reverse transcriptase inhibitors (NRTIs) ^[1]. Although initially believed to be an extremely rare event ^[2], recent studies demonstrated elevated serum lactate levels in asymptomatic patients receiving NRTIs ^[3-5], with an estimated incidence of between 15 and 35%. Without longitudinal studies it is not known whether severe NRTI-induced lactic acidosis (serum lactate level greater than 5 mmol/l, with a serum bicarbonate level lower than 20 mmol/l) is preceded by periods of mild hyperlactatemia (serum lactate levels between 2 and 5 mmol/l). It is possible that both represent the extremes of a single syndrome, because recent reports have appeared on patients with mild hyperlactatemia with symptoms similar to those of patients with lactic acidosis, including extreme fatigue, vomiting and abdominal pain ^[6,7].

In this issue of AIDS, John et al. ^[8] present the first, prospectively conducted, longitudinal study on the aspects of hyperlactatemia. Their study confirms the significant increase in serum lactate levels after the initiation of NRTIs, although it is unclear why patients who start NRTIs have lower lactate levels (0.74 versus 1.34 mmol/l) than antiretroviral-naive patients. An important observation in their study is the finding that lactate levels seem to stabilize for a prolonged period after the initial increase. In the one patient developing lactic acidosis, with previous lactate values, no gradual increase was observed. The authors therefore conclude that severe lactic acidosis can develop without previous, asymptomatic hyperlactatemia. Unfortunately, they do not emphasize this conclusion by commenting on the pattern of lactate increase in the five patients whose treatments were switched because of symptomatic hyperlactatemia.

The frequently cited estimated frequency of lactic acidosis (1.2 cases/1000 person years) by Fortgang et al. ^[2] was established in a period when NRTI monotherapy was the standard and clearly needs adjustment; in the study by John et al. ^[8] the incidence was 3.9/1000 person years. If the five cases of symptomatic hyperlactatemia are included as a prelude to severe lactic acidosis, the frequency would increase to 13.6/1000 patient years, which corresponds surprisingly well to the figure estimated by Lonergan et al. ^[9] for symptomatic hyperlactatemia (14.5/1000 patient years). Although this incidence is still not high, it appears to be increasing. Whether this is caused by an intensification of antiretroviral therapy or the use of more toxic NRTIs (such as stavudine) is not clear.

One of the most important messages from the study by John et al. ^[8] was the fact that mild hyperlactatemia, found in asymptomatic patients, appears to have a poor positive predictive value for the development of lactic acidosis. That study does not therefore endorse routine measurement in all NRTI-treated patients, which is in accordance with our own observations ^[5].

When then should lactate be measured? Currently, it seems prudent to advise lactate measurement only in those patients presenting with extreme fatigue, sudden weight loss, (vague) abdominal pain, unexplained nausea and vomiting, or sudden dyspnoea. For the interpretation of a lactate value, the attached algorithm can be followed (Fig. 1). The interruption of NRTIs remains the safest option in cases of severe hyperlactatemia and lactic acidosis, whether or not combined with supportive treatments (such as riboflavin or l-carnithine) ^[10]. Changing NRTIs (such as switching from stavudine to abacavir) might be a reasonable alternative in less severe cases, although this will not be feasible for every patient. Similar switches can be tried during cautious rechallenging after a temporary interruption ^[9]. Using these procedures, symptomatic hyperlactatemia may become a feature that can be handled, although these strategies need to be confirmed in properly conducted, prospective studies. Meanwhile, routine lactate measurements in asymptomatic patients should be avoided, because they do not provide predictive information, and could lead to unnecessary treatment alterations.

Fig. 1