Skip Navigation

Hyperlactatemia and Hepatic Abnormalities in 10 Human Immunodeficiency Virus-Infected Patients Receiving Nucleoside Analogue Combination Regimens

  1. J. Tyler Lonergan,
  2. Cynthia Behling,
  3. Hans Pfander,
  4. Tarek I. Hassanein, and
  5. W. Christopher Mathews
  1. University of California San Diego Medical Center
  1. Reprints or correspondence: Dr. J. Tyler Lonergan, University of California San Diego Medical Center, 200 West Arbor Dr., San Diego, CA 92103-8681 (tlonergan{at}ucsd.edu).
 
Next Section

Abstract

During a 6-and-a-half month period, we identified 10 human immunodeficiency virus (HIV)-infected men who were receiving antiretroviral regimens, including nucleoside analogues, and who developed unexplained reproducible hyperlactatemia in association with either abdominal symptoms or an unaccounted-for elevated alanine aminotransferase level, or both. After careful consideration of the possible etiologies, antiretrovirals were discontinued; lactate levels normalized in all patients. The estimated incidence of this phenomenon in our clinic was 20.9 cases per 1000 person-years of nucleoside analogue treatment. These observations extend the spectrum of the nucleoside analogue-induced lactic acidosis/hepatic steatosis syndrome by the identification of a subtle and perhaps earlier form, which has characteristic symptoms and laboratory abnormalities, and a favorable prognosis on discontinuation of antiretroviral therapy.

An often fatal syndrome of lactic acidosis and hepatic steatosis is a recognized but rare complication of nucleoside analogue reverse transcriptase inhibitor (NRTI) therapy for HIV infection [1]. Mitochondrial toxicity is the proposed mechanism of hepatic injury [1]. Nucleoside analogues can also inhibit DNA polymerase-γ, the enzyme responsible for mitochondrial DNA synthesis [2]. The ensuing mitochondrial dysfunction may result in lactic acidosis, hepatic steatosis, and other adverse events (i.e., myopathy, pancreatitis, and peripheral neuropathy) that are observed in patients treated with this class of antiretrovirals [2, 3].

From mid-July 1998 through January 1999, we identified 10 adult HIV-infected patients who attended an urban, university-based HIV clinic and had a milder and possibly earlier form of the lactic acidosis/hepatic steatosis syndrome. These patients were receiving antiretroviral therapy, including ≥2 nucleoside analogues, and had unexplained reproducible hyperlactatemia associated with either abdominal symptoms or an unaccounted-for elevated alanine aminotransferase (ALT) level, or both. The first case will be described in detail, and features of the remaining cases will be summarized.

Previous SectionNext Section

Case Report

A 36-year-old homosexual man with HIV infection since 1991 (CD4+ cell count, 652 cells/mm3; HIV-1 RNA viral load, 23,758 copies/mL by use of the Amplicor PCR assay [Roche, Nutley, NJ]) presented for evaluation of a 1-week history of worsening abdominal pain and distention, nausea, and diarrhea. Nine months before presentation, the patient had initiated a salvage antiretroviral regimen that included ritonavir 400 mg twice daily, soft-gel saquinavir 400 mg twice daily, stavudine 40 mg twice daily, and didanosine 300 mg once daily. He had previously been treated with the antiretrovirals indinavir, nelfinavir, zidovudine, and lamivudine. His only other medication was trimethoprim-sulfamethoxazole (TMP-SMZ), which was prescribed because of a CD4+ cell count nadir of 41 cells/mm3.

Upon examination, the patient was overweight and in mild distress with a temperature of 36.4°C and a blood pressure of 136/74 mm Hg. The remainder of the examination findings were notable only for abdominal distention and right upper quadrant tenderness.

Abnormal laboratory findings included the following: aspartate aminotransferase (AST) level, 139 U/L (normal range, 10–45 U/L); ALT level, 211 U/L (normal range, 10–45 U/L); bicarbonate level, 17 mmol/L (normal range, 24–31 mmol/L); anion gap, 22 (normal range, 10–12 mmol/L); venous plasma lactate level, 5.3 mmol/L (normal range, 0.7–2.1 mmol/L); and an arterial pH of 7.43, a pO2 level of 87 mm Hg, and a pCO2 level of 30 mm Hg. Results of stool studies, pancreatic enzyme, bilirubin, and alkaline phosphatase levels, complete blood count, and viral hepatitis serologies were unremarkable.

Abdominal ultrasound showed mild hepatomegaly (16.7 cm in length, normal 15 cm) with normal echogenicity. Evaluation of a liver biopsy specimen revealed mild macrovesicular steatosis and severe diffuse microvesicular steatosis. Oil-red-O staining of frozen biopsy tissue confirmed the steatosis. There was no significant inflammation, fibrosis, or hepatocyte necrosis. Electron microscopic examination demonstrated increased density in the inner mitochondrial compartment without swelling or disintegration of the mitochondria.

Antiretroviral medications were suspended 2 days after presentation. Twelve days later the patient's condition had improved markedly; his lactate level was 4.0 mmol/L and his ALT levels was 106 U/L. Within 34 days the patient's symptoms resolved and his laboratory values normalized.

Previous SectionNext Section

Results

During a 6-and-a-half month period, 10 patients who were receiving nucleoside analogues were identified with hyperlactatemia and either abdominal symptoms or an abnormal ALT level, or both. Table 1 summarizes the pertinent characteristics of these patients. All patients were adult men (median age, 35 years; range, 25–48 years), and the majority were overweight (median body mass index [BMI], 28.5 kg/m2; range, 20.1–37.5 kg/m2). Their median CD4+ cell count was 375 cells/mm3 (range, 217–1397 cells/mm3), and their median plasma HIV-1 RNA viral load was 780 copies/mL (range, <40–27,941 copies/mL). All patients were taking stavudine and ≥1 other nucleoside analogue, and 8 were also receiving either a protease inhibitor or a nonnucleoside reverse transcriptase inhibitor, or both, with median treatment duration of 10 months (range, 4–20 months).

Figure 1
View larger version:
Figure 1

Masson trichome-stained section of liver shows mixed micro- and macrovesicular steatosis. Arrow, Liver cell containing small cytoplasmic fat droplets (original magnification, ×400).

Table 1
View larger version:
Table 1

Characteristics of 10 HIV-infected patients with nucleoside-analogue induced hyperlactatemia (HL) and hepatic abnormalities.

Three patients had concomitant chronic hepatitis B or C. No patient had concurrent clinical disorders that would either predispose to or result in lactic acidosis. In addition, none of the patients reported excessive alcohol ingestion or strenuous physical activity during the previous 6 months.

Gastrointestinal complaints prompted evaluation in 8 patients (including all 3 patients with chronic viral hepatitis), whereas ALT level elevations not attributable to another cause led to the identification of hyperlactatemia in 2 asymptomatic patients. The most common symptoms were abdominal pain (n=8), nausea (7), and distention (6). Before the onset of these symptoms, the patients had been tolerating their medications well for several months.

All patients had reproducible elevated venous lactate levels (median peak, 4.4 mmol/L; range, 2.9–6.2 mmol/L) while receiving antiretroviral therapy. To minimize false lactate elevations, blood was collected by venipuncture into a tube containing sodium fluoride (an inhibitor of lactate dehydrogenase) and immediately placed on ice. The iced specimen was then centrifuged, and the plasma lactate level was measured by using a Beckman Synchron CX7Δ (Beckman Instruments, Brea, CA) instrument. If the lactate level was elevated, it was confirmed by having the patient return to the clinic and repeating the measurement using the same technique. Seven of 10 patients with hyperlactatemia had evidence of metabolic acidosis—either decreased bicarbonate or increased unmeasured anions—but only 2 of these patients had both simultaneously.

Median peak ALT level was 3 times the upper limit of normal (range, 2.1–10.7 times normal). All the patients, with the exception of the 3 who had chronic viral hepatitis, had normal ALT levels prior to starting their current antiretroviral regimen. CT and/or ultrasonographic imaging of the liver was performed for 7 patients. Hepatomegaly, determined by use of ultrasound (>15 cm in length) was evident in 4 of 5 patients (median, 18.5 cm; range, 12.4–19.0 cm), and imaging by either technique was consistent with fatty infiltration in 5 patients.

To rule out other causes of liver disease and to provide supporting evidence, liver biopsies were performed for 6 patients. One patient's specimen showed changes consistent with moderately active hepatitis B. For the remaining 5 patients, biopsy specimens revealed a spectrum of similar findings including normal liver architecture and minimal portal inflammation. Bile ducts, arteries, and portal veins were normal. The hepatocytes showed variable degrees of micro- and macrovesicular steatosis (figure 1). The microvesicular fat involved 0–100% of liver cells, and the macrovesicular fat appeared in <2%–100% of liver cells. The biopsy specimen from a patient with 100% macrovesicular fat had features of steatohepatitis. In all of the biopsy specimens, the uninvolved liver cells had abundant eosinophilic granular cytoplasm. There were Kupffer cell aggregates containing brown-gray pigment and small fat vacuoles. With the exception of the biopsy specimen that resembling steatohepatitis, there was little hepatocyte necrosis. There was no iron in any of the specimens. Two specimens demonstrated glassy hepatocyte cytoplasm and hepatocyte rosettes. The patient without hepatic steatosis had been off antiretrovirals for 3 weeks before the biopsy. The other 5 patients' biopsies were performed within 13 days of drug discontinuation.

Electron microscopy available for 3 biopsy specimens showed lipid inclusions and prominent endoplasmic reticulum. One specimen had increased density of the inner mitochondrial membrane, whereas another had enlarged mitochondria containing paracrystalline inclusions and dense bodies.

All patients had favorable outcomes: only 1 required hospitalization, and none died. Clinical symptoms abated more rapidly than did biochemical markers. Within a few days of discontinuing their medications, most patients noticed significant symptomatic improvement. Lactate levels, in contrast, often increased initially after ceasing NRTI therapy, but in all cases were declining by 4 weeks. Median time to normalization of the lactate level was 62 days (range, 16–111 days).

Of 1245 patients seen in our clinic from mid-July 1998 through the end of January 1999, the overall incidence was 14.8/1000 person-years, and among those receiving nucleoside analogue therapy (n=883), 20.9/1000 person-years.

Previous SectionNext Section

Discussion

Our observations extend the spectrum of the lactic acidosis/hepatic steatosis syndrome by the discovery of a milder and possibly earlier presentation. This variant has characteristic symptoms of abdominal pain, nausea, and distention, and laboratory abnormalities of slightly elevated lactate and ALT levels, as well as clinical and biochemical resolution with discontinuation of antiretroviral therapy. We ascribe the hyper-lactatemia and hepatic abnormalities found in our patients to nucleoside analogue toxicity for the following reasons: (1) these were the only medications common to all patients that are known to cause lactic acidosis; (2) the symptoms and laboratory abnormalities abated once antiretrovirals were discontinued; and (3) other possible explanations were excluded. The common biochemical and pathologic findings suggest a relationship to the previously described syndrome. Because most of the patients were also receiving protease inhibitors or nonnucleoside reverse transcriptase inhibitors, neither class of which has been associated with lactic acidosis, we cannot exclude a metabolic interaction among these drug classes.

Similar to other reports describing the lactic acidosis/hepatic steatosis syndrome, our patients had received nucleoside analogue therapy for several months before presentation, and most of them were overweight, had gastrointestinal complaints, hepatomegaly, slightly elevated aminotransaminases, and evidence of hepatic steatosis, either by imaging or biopsy [1, 4–21]. In contrast, our patients had much milder elevations in lactate levels. Only 4 of our 10 patients would have met the usual criterion for lactic acidosis (lactate level, >5 mmol/L) [22]. In accordance with this definition, we describe the abnormalities we observed as hyperlactatemia rather than as lactic acidosis. In addition, the majority of our patients had little evidence of metabolic acidosis, according to bicarbonate and anion gap measurements. Therefore, these markers cannot be solely relied on to screen for the milder syndrome.

Our patients also had better outcomes than did those patients described in previous studies. There were no deaths and only 1 hospitalization among our 10 patients. Previous series have reported a mortality rate >50% [1, 5, 10, 16]. Of 36 cases reported in the literature, the majority required hospitalization, and two-thirds died [4–21]. These higher morbidity and mortality rates may be from spectrum bias resulting in identification of only the most severe cases.

The incidence of this milder syndrome appears to be higher than reported previously. In a patient population similar to our own, Fortgang et al. [8] identified 2 cases, both fatal, over a 5-year period with an overall estimated incidence of 0.9/1000 person-years and 1.3/1000 person-years of nucleoside analogue exposure. The incidence rates in our series were ∼16-fold greater than these estimates. Our rates are conservative, because lactate levels were probably not obtained in every patient taking nucleoside analogues who presented to the clinic with either abdominal symptoms or an unexplained abnormal ALT level during this period.

The higher incidence in the present report may be a result of broadened case definition. Fortgang et al. [8] included only severe cases of lactic acidosis in their incidence estimates. Alternatively, these cases may reflect an increasing incidence caused by changes in treatment strategy. Previous rates were estimated on the basis of the use of nucleoside analogue monotherapy. Since the dissemination of clinical trial results documenting improved outcomes with combination antiretroviral therapy, dual nucleoside analogue combinations (in conjunction with protease inhibitors or nonnucleoside reverse transcriptase inhibitors) are increasingly prevalent and may result in greater cumulative exposure to the potential mitochondrial toxicity of this drug class [23, 24].

Another possible basis for the higher incidence is the increased use of stavudine. Previous rate estimates were performed before widespread use of this nucleoside analogue. Stavudine, a regimen component in all our cases, has been shown in vitro to be the most damaging to mitochondria among the frequently prescribed nucleoside analogues [25–27]. Recently, other investigators have implicated nucleoside analogues (especially stavudine) in the pathogenesis of HIV-associated lipo-dystrophy and have identified hyperlactatemia as a marker of mitochondrial toxicity in that setting [28].

We do not know whether the syndrome we have described would have progressed to the more severe form if treatment had been continued. However, recognition of the milder variant may simplify the diagnostic evaluation of patients receiving nucleoside analogue-containing regimens who present with gastrointestinal complaints and/or unexplained hepatic abnormalities. Many patients in our series underwent several nondiagnostic studies before an association between the nucleoside analogues and the symptoms and laboratory abnormalities was made. We recommend that providers consider obtaining a lactate level for any patient receiving nucleoside analogues who presents with ≥1 of the following: abdominal symptoms, metabolic acidosis, or elevated aminotransaminases. We consider it prudent to discontinue nucleoside analogue therapy for these patients when elevation of the lactate level, not attributable to another cause, is confirmed. The routine measurement of lactate in all patients receiving nucleoside analogues in an attempt to provide early screening and detection of the syndrome cannot be supported at this time.

Additional research is needed to address several questions raised by these observations. First, in the era of multinucleoside analogue combination regimens, what is the current incidence of hyperlactatemia and hepatic abnormalities, and does it differ by regimen? Second, what is the prognosis for patients with the milder syndrome we have described, both with and without treatment interruption? Third, is rechallenge with the same or other members of the nucleoside analogue class safe if no other treatment options are available for the underlying HIV infection? Lastly, can risk factors and preventive factors for mitochondrial toxicity be identified, so that patients can experience the proven benefit of nucleoside analogue therapy without risk of hepatic damage?

Previous SectionNext Section

Footnotes

  • See editorial response by Brinkman on pages 167–9.

  • Received July 29, 1999.
  • Revision received December 23, 1999.
Previous Section
 

References

  1. ↵
    1. Stein DS
    . A new syndrome of hepatomegaly with severe steatosis in HIV-seropositive patients. AIDS Clin Care 1994;6:17-21.
  2. ↵
    1. Parker WP,
    2. Cheng YC
    . Mitochondrial toxicity of antiviral nucleoside analogs. J NIH Research 1994;6:57-61.
  3. ↵
    1. Brinkman K,
    2. ter Hofstede HJM,
    3. Burger DM,
    4. Smeitink J,
    5. Koopmans PP
    . Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS 1998;12:1735-44.
    MedlineWeb of Science
  4. ↵
    1. Lai KK,
    2. Gang DL,
    3. Zawacki JK,
    4. Cooley TP
    . Fulminant hepatic failure associated with 2',3'-dideoxyinosine (ddI). Ann Intern Med 1991;115:283-4.
    Abstract/FREE Full Text
  5. ↵
    1. Freiman JP,
    2. Helfert KE,
    3. Hamrell MR,
    4. Stein DS
    . Hepatomegaly with severe steatosis in HIV-seropositive patients. AIDS 1993;7:379-85.
    MedlineWeb of Science
    1. Bissuel F,
    2. Bruneel F,
    3. Habersetzer F,
    4. et al
    . Fulminant hepatitis with severe lactic acidosis in HIV-infected patients on didanosine therapy. J Intern Med 1994;235:367-71.
    MedlineWeb of Science
    1. Olano JP,
    2. Borucki MJ,
    3. Wen JW,
    4. Haque AK
    . Massive hepatic steatosis and lactic acidosis in a patient with AIDS who was receiving zidovudine. Clin Infect Dis 1995;21:973-6.
    Abstract/FREE Full Text
  8. ↵
    1. Fortgang IS,
    2. Belitsos PC,
    3. Chaisson RE,
    4. Moore RD
    . Hepatomegaly and steatosis in HIV-infected patients receiving nucleoside analog antiretroviral therapy. Am J Gastroenterol 1995;90:1433-6.
    MedlineWeb of Science
    1. Lenzo NP,
    2. Garas BA,
    3. French MA
    . Hepatic steatosis and lactic acidosis associated with stavudine treatment in an HIV patient: a case report [letter]. AIDS 1997;11:1294-6.
    MedlineWeb of Science
  10. ↵
    1. Sundar K,
    2. Suarez M,
    3. Banogon PE,
    4. Shapiro JM
    . Zidovudine-induced fatal lactic acidosis and hepatic failure in patients with acquired immunodeficiency syndrome: report of two patients and review of the literature. Crit Care Med 1997;25:1425-30.
    CrossRefMedlineWeb of Science
    1. Fouty B,
    2. Frerman F,
    3. Reves R
    . Riboflavin to treat nucleoside analogue-induced lactic acidosis [letter]. Lancet 1998;352:291-2.
    CrossRefMedlineWeb of Science
    1. Brinkman K,
    2. Ter Hofstede H,
    3. Veerkamp MJ,
    4. Kloke HJ,
    5. Willems JL,
    6. Wesseling P
    . Proceedings and abstracts of the 12th International Conference AIDS (Geneva, June 1998). Geneva: Marathon Multimedia; 1998. Fatal lactic acidosis following HAART containing stavudine (d4T), lamuvidine (3TC) and saquinavir [abstract 60998].
    1. Rubin DS,
    2. Izhar T,
    3. Qavi A
    . Proceedings and abstracts of the 12th International Conference AIDS (Geneva, June 1998). Geneva: Marathon Multimedia; 1998. The 'steatosis-lactic acidosis syndrome' (SLAS) in early stage HIV infection treated with HAART [abstract 60611].
    1. Chen SCA,
    2. Barker SM,
    3. Mitchell DH,
    4. Stevens SMB,
    5. O'Neill P,
    6. Cunningham AL
    . Concurrent zidovudine-induced myopathy and hepatotoxicity in patients treated for human immunodeficiency virus (HIV) infection. Pathology 1992;24:109-111.
    Medline
    1. Gradon JD,
    2. Chapnick EK,
    3. Sepkowitz DV
    . Zidovudine-induced hepatitis. J Intern Med 1992;231:317-18.
    MedlineWeb of Science
  16. ↵
    1. Chattha G,
    2. Arieff AI,
    3. Cummings C,
    4. Tierney LM Jr.
    . Lactic acidosis complicating the acquired immunodeficiency syndrome. Ann Intern Med 1993;118:37-9.
    Abstract/FREE Full Text
    1. Baram D,
    2. Cooke J
    . Lactic acidosis and AIDS [letter]. Ann Intern Med 1993;119:345.
    MedlineWeb of Science
    1. Aggarwal A,
    2. al-Talib K,
    3. Alabrash M
    . Type B lactic acidosis in an AIDS patient treated with zidovudine. Md Med J 1996;45:929-31.
    Medline
    1. Charton-Bain MC,
    2. Flamant M,
    3. Aubertin JM,
    4. et al
    . Lactic acidosis and hepatic mitochondrial changes during a treatment with zidovudine. Gastroenterol Clin Biol 1997;21:979-81.
    MedlineWeb of Science
    1. Chariot P,
    2. Drogou I,
    3. de Lacroix-Szmania I,
    4. et al
    . Zidovudine-induced mitochondrial disorder with massive liver steatosis, myopathy, lactic acidosis, and mitochondrial DNA depletion. J Hepatol 1999;30:156-60.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Acosta BS,
    2. Grimsley EW
    . Zidovudine-associated type B lactic acidosis and hepatic steatosis in an HIV-infected patient. South Med J 1999;92:421-3.
    MedlineWeb of Science
  22. ↵
    1. Madias NE
    . Lactic acidosis. Kidney Int 1986;29:752-74.
    MedlineWeb of Science
  23. ↵
    1. Hammer SM,
    2. Katzenstein DA,
    3. Hughes MD,
    4. et al
    . A trial comparing nucleoside monotherapy with combination therapy in HIV-infected adults with CD4 cell counts from 200 to 500 per cubic millimeter. N Engl J Med 1996;335:1081-90.
    CrossRefMedlineWeb of Science
  24. ↵
    Delta Coordinating Committee. Delta: a randomized double-blind controlled trial comparing combinations of zidovudine plus didanosine or zalcitabine with zidovudine alone in HIV-infected individuals. Lancet 1996;348:283-91.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Chen CH,
    2. Vazquez-Padua M,
    3. Cheng YC
    . Effect of anti-human immunodeficiency virus nucleoside analogs on mitochondrial DNA and its implication for delayed toxicity. Mol Pharmacol 1991;39:625-8.
    Abstract
    1. Bridges EG,
    2. Dutschman GE,
    3. Gullen EA,
    4. Chen YC
    . Favorable interaction of β-L(-) nucleoside analogues with clinically approved anti-HIV nucleoside analogues for the treatment of human immunodeficiency virus. Biochem Pharmacol 1996;51:731-6.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Parker WB,
    2. Shaddix SC,
    3. Vince R,
    4. Bennett LL Jr.
    . Lack of mitochondrial toxicity in CEM cells treated with carbovir. Antiviral Res 1997;34:131-6.
    CrossRefMedlineWeb of Science
  28. ↵
    1. Carr A,
    2. Miller J,
    3. Law M,
    4. Cooper DA
    . A syndrome of lipodystrophy (LD), lactic acidaemia and liver dysfunction associated with HIV nucleoside analogue reverse transcriptase inhibitor therapy: contribution to PI- related LD syndrome [abstract 011]. AIDS 2000;14:F25-32.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2000) 31 (1): 162-166. doi: 10.1086/313912
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most