Hematologic Effects of Antimicrobials: Focus on the Oxazolidinone...
Hematologic Effects of Antimicrobials: Focus on the Oxazolidinone...
The oxazolidinone linezolid (Zyvox, Pharmacia Corp., Peapack, NJ) recently was approved for clinical use in a variety of gram-positive infections including those due to methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecium (VRE). It is the first representative of a truly novel class and was shown in a series of multinational clinical trials to be efficacious and safe in serious or problematic infections for which there are few therapeutic alternatives.
Although myelosuppression has been noted with linezolid, recently, three cases of linezolid-associated reversible bone marrow suppression with anemia, reticulocytopenia, and thrombo-cytopenia were reported. The report prompted our review of the hematologic effects of linezolid.
Many antimicrobial agents affect the hematopoietic system. The most widely known effect is chloramphenicol-induced, nonreversible aplastic anemia, which results from 1/40-60,000 administrations. The exact rates of hematologic events for other agents are not easy to ascertain, but the following rates are reported in package inserts: less than 1% of patients taking amoxicillin-clavulanic acid experience anemia, thrombocytopenia, and leukopenia; and ceftriaxone leads to eosinophilia and leukopenia in 6% and 2.1% of patients, respectively, and to anemia, neutropenia, and thrombocytopenia, each in less than 1% of patients. Cefuroxime is associated with decreases in hemoglobin and hematocrit in approximately 1/10 patients, whereas 1/750 may develop leukopenia. Levofloxacin is linked to thrombocytopenia and anemia in less than 1/200 patients, and vancomycin causes thrombocytopenia in less than 1/1000 patients. Leukopenia is associated with several ß-Lactams including nafcillin, oxacillin, piperacillin, cephalexin, imipenem, and aztreonam. Most of the hematologic effects are dose and duration dependent and reversible on drug discontinuation.
Various mechanisms of hematopoietic toxicity include direct cytolysis, immune-mediated destruction, genotoxicity, and mitochondrial protein synthesis inhibition. Antimicrobial hematopoietic suppression clearly can be harmful, but usually it is reversible, easily monitored, and thus clinically manageable.
Typically, hematopoietic suppression is manifested in three ways: thrombocytopenia, anemia, and neutropenia. Thrombocytopenia can be due to marrow suppression, immune-mediated destruction, or platelet aggregation. It is defined as a decrease in the number of platelets to below 150 x 10/mm (normal range 150-450 x 10/mm). Many antimicrobial agents are associated with thrombocytopenia, including sulfonamides, penicillins, cephalosporins, vancomycin, and trimethoprim.
Anemia is a condition in which the blood is deficient in red blood cells, as measured by the red blood cell count, hemoglobin, or hematocrit. Typically the red blood cell count is 4.6-6.2 x 10/mm for men and 4.2-5.4 x 10/mm for women; hemoglobin is 13.5-18.0 g/dl for men and 12.0-16.0 g/dl for women. Anemias induced by drugs are immune-mediated hemolytic, reversible myelosuppressive, and aplastic. The latter is postulated to be due to genotoxicity. One researcher demonstrated that chloramphenicol caused direct damage to human lymphocytes, leading to nonreplication, but this effect was not seen with thiamphenicol or linezolid. Clearly molecular structure plays a role in this serious adverse effect.
Drug-induced neutropenia is caused by a range of therapeutic agents. Associated antimicrobial agents include ß-Lactams, vancomycin, chloram-phenicol, ciprofloxacin, nitrofurantoin, and several antituberculous agents. Direct bone marrow suppression, immune-mediated damage to precursor cells, and destruction and clearance of neutrophils are all associated with drug-induced neutropenia.
Linezolid has excellent antimicrobial activity against both susceptible and multiresistant strains of gram-positive species. Over 2000 patients were treated with linezolid in clinical trials designed to evaluate gram-positive activity in complicated skin and soft tissue infections, and in community- and hospital-acquired pneumonias, including those caused by MRSA and VRE. Investigators predicated the length of therapy by specific study protocol criteria. The majority of patients received linezolid 600 mg every 12 hours for 10-14 days, with the exception of patients with suspected or documented VRE infection who were treated for up to 28 days (Table 1).
Patients underwent laboratory evaluations at study entry and at three subsequent occasions during the study. Hematologic values -- hemoglobin, platelet counts, and white and red blood cell counts -- were compared for linezolid with various comparator agents: ceftriaxone/cefpodoxime, oxacillin/dicloxacillin, or vancomycin. No statistically significant differences were observed (Table 2). Subgroup analysis of the extent of any changes, such as comparisons of mild or severe shifts in blood counts, did not show any significant differences. For linezolid and comparators, respectively, the overall rates of thrombocytopenia were 2.4% and 1.5%, and anemia 0.7% and 0.2%.
In the first 6 months of approved use, 55,000 patients were treated. Spontaneous postmarketing surveillance recorded 72 cases of hematologic abnormalities: thrombocytopenia 32 cases, anemia 19, pancytopenia 13, leukopenia 2, and complex disorders (e.g, bicytopenia) 5 (data on file, Pharmacia Corp., Peapack, NJ). Overall, hematologic events were observed in approxi-mately 1/750 exposed patients. Specifically thrombocytopenia occurred in 1/1700 patients, anemia in 1/2900, and pancytopenia in 1/6000. These rates compare favorably with those described earlier with other commonly prescribed antimicrobial agents such as amoxicillin-clavulanic acid or ceftriaxone. The spontaneous reports regarding linezolid were fully pursued to establish all of the key details and potential causality. Table 3 outlines the clinical significance criteria for determining abnormal hematologic events.
Of the 32 cases of thrombocytopenia, 15 were reported as serious (defined as requiring medical intervention) in 9 women and 6 men aged 54-86 years (mean 71.0 yrs); of these 15 assessable cases, 6 patients were diabetic. The duration of therapy was 7-28 days for bacteremia (5 cases), endocarditis (2), osteomyelitis (2), nonspecific systemic infections (2), and other infections (4).
The median time to onset of thrombocytopenia was 17 days, with a mean of 31.7 days (range 2-180 days). Of these cases only two occurred during the first week of therapy, four during the second week, and all other instances appeared between the third and sixth weeks. All of the thrombocytopenic patients were taking several agents for a variety of underlying medical conditions, four of whom had prior thrombocytopenia (platelet counts 70-90 x 10/mm, 57 x 10/mm, 132 x 10/mm, and 133 x 10/mm [normal 150-450 x 10/mm]). Platelet counts recovered in nine patients after discontinuing linezolid. Three patients died while thrombocytopenic, but due to other causes: sepsis and pulmonary embolus; sepsis, renal failure, and chronic heart failure; and sepsis with disseminated intravascular coagulation overlying malignancy, diabetes, and hypertension. Two patients suffered gastrointestinal and adrenal hemorrhagic episodes; however, both had platelet counts between 100 x 10/mm and 150 x 10/mm.
Anemia was reported in 19 patients, and 12 of them were serious (seven men, five women, aged 45-79 yrs [mean 59.9 yrs]). Anemia developed in 7-90 days (mean 42.2 days), with five patients receiving at least 6 weeks of linezolid. All but two patients made a full recovery after stopping the drug. A 51-year-old woman with chronic osteomyelitis received linezolid for almost 7 weeks, at which time therapy was discontinued due to a white blood cell count of 2.6 x 10/mm and hemoglobin of 7.8 g/dl.
Pancytopenia, defined as a decline in the red blood cells, white blood cells, and platelets, was reported in 13 patients, each of whom had a complex medical history. Twelve of 13 were recorded as serious events. The events occurred at 9-70 days of treatment (mean 33.3 days), and all but two occurred after 21 days. The outcome is known for eight patients: five recovered, two have not yet recovered, and one died. The patient who died had bone marrow suppression as well as diabetes, cardiovascular disease, and end-stage renal failure, all of which contributed to death. Of the two patients not yet recovered, one young woman received a bone marrow transplant, was taking 30 concomitant drugs, and had not engrafted by the time of investigation. The reporting physician considered the myelosuppression not due to linezolid. The other patient, before starting linezolid, had low hemoglobin, platelet, and leukocyte counts, probably due to prior vancomycin use. After discontinuing the drug, the patient was stable and the counts remained low.
Bicytopenia is a condition in which two hematologic abnormalities, for example, thrombocytopenia and anemia, are observed simultaneously. This was noted in five patients who received linezolid for 24-60 days for severe deep-seated resistant gram-positive infections. Two patients experienced anemia and neutropenia, two had anemia and thrombocytopenia, and one had thrombocytopenia and neutropenia. Each patient had taken or was receiving other agents known to cause myelosuppression including other antimicrobial agents such as vancomycin or ciprofloxacin. In all five cases, there were no other clinically significant adverse events, and in each instance discontinuation of linezolid led to reversal of the effects.
Several etiologies may exist for the hematologic abnormalities reported. Most patients in whom the events were observed had longstanding, severe infections and had taken many antibiotics. Severe infections alone, independent of therapy, may be associated with bone marrow depression. An equal number of patients had severe underlying diseases that may cause hematologic abnormalities. One of the 32 thrombocytopenic patients had undergone bone marrow trans-plantation, which often leads to diminished marrow reserves. Finally, most patients were on concomitant agents, some of which are associated with hematologic effects.
Evaluation by a panel of experts of the spontaneous reports and the clinical trial database suggested that the linezolid-related hematologic effects were mild to moderate in severity, with reversible anemias or cytopenias usually after 4 weeks of treatment. In patients whose outcomes are known, the affected parameters returned to pretreatment values when linezolid was discontinued. Prolonged thrombocytopenia or anemia associated with linezolid occurred after treatment beyond 14 days. The temporal and reversible nature of the hematologic events suggests transient bone marrow suppression. These findings eliminate any hemolytic process or aplastic anemia, both of which are linked to fatal consequences with other antimicrobial agents such as chloramphenicol.
After review of these cases, it is evident that linezolid was not the sole cause of the hematologic findings; they may be related to underlying disease or concomitant agents. Many patients had preexisting cytopenias or decreased bone marrow reserves. As such, these patients might be more sensitive to any inhibitory effect of linezolid on the bone marrow. Nevertheless, in cases where outcomes are known, the effects were reversible. Patients should be monitored appropriately, especially those with preexisting bone marrow suppression or diminished bone marrow reserves. For those being treated for longer than 14 days, the clinical benefit of linezolid in serious gram-positive infections should be weighed against potential, but reversible, hematologic effects.
The oxazolidinone linezolid (Zyvox, Pharmacia Corp., Peapack, NJ) recently was approved for clinical use in a variety of gram-positive infections including those due to methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecium (VRE). It is the first representative of a truly novel class and was shown in a series of multinational clinical trials to be efficacious and safe in serious or problematic infections for which there are few therapeutic alternatives.
Although myelosuppression has been noted with linezolid, recently, three cases of linezolid-associated reversible bone marrow suppression with anemia, reticulocytopenia, and thrombo-cytopenia were reported. The report prompted our review of the hematologic effects of linezolid.
Many antimicrobial agents affect the hematopoietic system. The most widely known effect is chloramphenicol-induced, nonreversible aplastic anemia, which results from 1/40-60,000 administrations. The exact rates of hematologic events for other agents are not easy to ascertain, but the following rates are reported in package inserts: less than 1% of patients taking amoxicillin-clavulanic acid experience anemia, thrombocytopenia, and leukopenia; and ceftriaxone leads to eosinophilia and leukopenia in 6% and 2.1% of patients, respectively, and to anemia, neutropenia, and thrombocytopenia, each in less than 1% of patients. Cefuroxime is associated with decreases in hemoglobin and hematocrit in approximately 1/10 patients, whereas 1/750 may develop leukopenia. Levofloxacin is linked to thrombocytopenia and anemia in less than 1/200 patients, and vancomycin causes thrombocytopenia in less than 1/1000 patients. Leukopenia is associated with several ß-Lactams including nafcillin, oxacillin, piperacillin, cephalexin, imipenem, and aztreonam. Most of the hematologic effects are dose and duration dependent and reversible on drug discontinuation.
Various mechanisms of hematopoietic toxicity include direct cytolysis, immune-mediated destruction, genotoxicity, and mitochondrial protein synthesis inhibition. Antimicrobial hematopoietic suppression clearly can be harmful, but usually it is reversible, easily monitored, and thus clinically manageable.
Typically, hematopoietic suppression is manifested in three ways: thrombocytopenia, anemia, and neutropenia. Thrombocytopenia can be due to marrow suppression, immune-mediated destruction, or platelet aggregation. It is defined as a decrease in the number of platelets to below 150 x 10/mm (normal range 150-450 x 10/mm). Many antimicrobial agents are associated with thrombocytopenia, including sulfonamides, penicillins, cephalosporins, vancomycin, and trimethoprim.
Anemia is a condition in which the blood is deficient in red blood cells, as measured by the red blood cell count, hemoglobin, or hematocrit. Typically the red blood cell count is 4.6-6.2 x 10/mm for men and 4.2-5.4 x 10/mm for women; hemoglobin is 13.5-18.0 g/dl for men and 12.0-16.0 g/dl for women. Anemias induced by drugs are immune-mediated hemolytic, reversible myelosuppressive, and aplastic. The latter is postulated to be due to genotoxicity. One researcher demonstrated that chloramphenicol caused direct damage to human lymphocytes, leading to nonreplication, but this effect was not seen with thiamphenicol or linezolid. Clearly molecular structure plays a role in this serious adverse effect.
Drug-induced neutropenia is caused by a range of therapeutic agents. Associated antimicrobial agents include ß-Lactams, vancomycin, chloram-phenicol, ciprofloxacin, nitrofurantoin, and several antituberculous agents. Direct bone marrow suppression, immune-mediated damage to precursor cells, and destruction and clearance of neutrophils are all associated with drug-induced neutropenia.
Linezolid has excellent antimicrobial activity against both susceptible and multiresistant strains of gram-positive species. Over 2000 patients were treated with linezolid in clinical trials designed to evaluate gram-positive activity in complicated skin and soft tissue infections, and in community- and hospital-acquired pneumonias, including those caused by MRSA and VRE. Investigators predicated the length of therapy by specific study protocol criteria. The majority of patients received linezolid 600 mg every 12 hours for 10-14 days, with the exception of patients with suspected or documented VRE infection who were treated for up to 28 days (Table 1).
Patients underwent laboratory evaluations at study entry and at three subsequent occasions during the study. Hematologic values -- hemoglobin, platelet counts, and white and red blood cell counts -- were compared for linezolid with various comparator agents: ceftriaxone/cefpodoxime, oxacillin/dicloxacillin, or vancomycin. No statistically significant differences were observed (Table 2). Subgroup analysis of the extent of any changes, such as comparisons of mild or severe shifts in blood counts, did not show any significant differences. For linezolid and comparators, respectively, the overall rates of thrombocytopenia were 2.4% and 1.5%, and anemia 0.7% and 0.2%.
In the first 6 months of approved use, 55,000 patients were treated. Spontaneous postmarketing surveillance recorded 72 cases of hematologic abnormalities: thrombocytopenia 32 cases, anemia 19, pancytopenia 13, leukopenia 2, and complex disorders (e.g, bicytopenia) 5 (data on file, Pharmacia Corp., Peapack, NJ). Overall, hematologic events were observed in approxi-mately 1/750 exposed patients. Specifically thrombocytopenia occurred in 1/1700 patients, anemia in 1/2900, and pancytopenia in 1/6000. These rates compare favorably with those described earlier with other commonly prescribed antimicrobial agents such as amoxicillin-clavulanic acid or ceftriaxone. The spontaneous reports regarding linezolid were fully pursued to establish all of the key details and potential causality. Table 3 outlines the clinical significance criteria for determining abnormal hematologic events.
Of the 32 cases of thrombocytopenia, 15 were reported as serious (defined as requiring medical intervention) in 9 women and 6 men aged 54-86 years (mean 71.0 yrs); of these 15 assessable cases, 6 patients were diabetic. The duration of therapy was 7-28 days for bacteremia (5 cases), endocarditis (2), osteomyelitis (2), nonspecific systemic infections (2), and other infections (4).
The median time to onset of thrombocytopenia was 17 days, with a mean of 31.7 days (range 2-180 days). Of these cases only two occurred during the first week of therapy, four during the second week, and all other instances appeared between the third and sixth weeks. All of the thrombocytopenic patients were taking several agents for a variety of underlying medical conditions, four of whom had prior thrombocytopenia (platelet counts 70-90 x 10/mm, 57 x 10/mm, 132 x 10/mm, and 133 x 10/mm [normal 150-450 x 10/mm]). Platelet counts recovered in nine patients after discontinuing linezolid. Three patients died while thrombocytopenic, but due to other causes: sepsis and pulmonary embolus; sepsis, renal failure, and chronic heart failure; and sepsis with disseminated intravascular coagulation overlying malignancy, diabetes, and hypertension. Two patients suffered gastrointestinal and adrenal hemorrhagic episodes; however, both had platelet counts between 100 x 10/mm and 150 x 10/mm.
Anemia was reported in 19 patients, and 12 of them were serious (seven men, five women, aged 45-79 yrs [mean 59.9 yrs]). Anemia developed in 7-90 days (mean 42.2 days), with five patients receiving at least 6 weeks of linezolid. All but two patients made a full recovery after stopping the drug. A 51-year-old woman with chronic osteomyelitis received linezolid for almost 7 weeks, at which time therapy was discontinued due to a white blood cell count of 2.6 x 10/mm and hemoglobin of 7.8 g/dl.
Pancytopenia, defined as a decline in the red blood cells, white blood cells, and platelets, was reported in 13 patients, each of whom had a complex medical history. Twelve of 13 were recorded as serious events. The events occurred at 9-70 days of treatment (mean 33.3 days), and all but two occurred after 21 days. The outcome is known for eight patients: five recovered, two have not yet recovered, and one died. The patient who died had bone marrow suppression as well as diabetes, cardiovascular disease, and end-stage renal failure, all of which contributed to death. Of the two patients not yet recovered, one young woman received a bone marrow transplant, was taking 30 concomitant drugs, and had not engrafted by the time of investigation. The reporting physician considered the myelosuppression not due to linezolid. The other patient, before starting linezolid, had low hemoglobin, platelet, and leukocyte counts, probably due to prior vancomycin use. After discontinuing the drug, the patient was stable and the counts remained low.
Bicytopenia is a condition in which two hematologic abnormalities, for example, thrombocytopenia and anemia, are observed simultaneously. This was noted in five patients who received linezolid for 24-60 days for severe deep-seated resistant gram-positive infections. Two patients experienced anemia and neutropenia, two had anemia and thrombocytopenia, and one had thrombocytopenia and neutropenia. Each patient had taken or was receiving other agents known to cause myelosuppression including other antimicrobial agents such as vancomycin or ciprofloxacin. In all five cases, there were no other clinically significant adverse events, and in each instance discontinuation of linezolid led to reversal of the effects.
Several etiologies may exist for the hematologic abnormalities reported. Most patients in whom the events were observed had longstanding, severe infections and had taken many antibiotics. Severe infections alone, independent of therapy, may be associated with bone marrow depression. An equal number of patients had severe underlying diseases that may cause hematologic abnormalities. One of the 32 thrombocytopenic patients had undergone bone marrow trans-plantation, which often leads to diminished marrow reserves. Finally, most patients were on concomitant agents, some of which are associated with hematologic effects.
Evaluation by a panel of experts of the spontaneous reports and the clinical trial database suggested that the linezolid-related hematologic effects were mild to moderate in severity, with reversible anemias or cytopenias usually after 4 weeks of treatment. In patients whose outcomes are known, the affected parameters returned to pretreatment values when linezolid was discontinued. Prolonged thrombocytopenia or anemia associated with linezolid occurred after treatment beyond 14 days. The temporal and reversible nature of the hematologic events suggests transient bone marrow suppression. These findings eliminate any hemolytic process or aplastic anemia, both of which are linked to fatal consequences with other antimicrobial agents such as chloramphenicol.
After review of these cases, it is evident that linezolid was not the sole cause of the hematologic findings; they may be related to underlying disease or concomitant agents. Many patients had preexisting cytopenias or decreased bone marrow reserves. As such, these patients might be more sensitive to any inhibitory effect of linezolid on the bone marrow. Nevertheless, in cases where outcomes are known, the effects were reversible. Patients should be monitored appropriately, especially those with preexisting bone marrow suppression or diminished bone marrow reserves. For those being treated for longer than 14 days, the clinical benefit of linezolid in serious gram-positive infections should be weighed against potential, but reversible, hematologic effects.
Source...