Ceftriaxone

* An immune complex mechanism for ceftriaxone sodiuminduced severe autoimmune hemolytic anemia has previously been demonstrated using routine blood bank techniques. We describe herein a patient with severe hemolysis that subsided once the drug was discontinued. Serologic techniques demonstrated immune complex-mediated ceftriaxone-dependent red cell antibodies. These findings were further supported by the results of flow cytometry, in which a change in basal red cell autofluorescence was seen in the presence of the antibody and the drug. Our case illustrates the adjunctive value of flow cytometry in the diagnosis of ceftriaxone-dependent red cell antibody.

(Arch Pathol Lab Med. 2004;128:905-907)

In 1991, the first fatal case of ceftriaxone sodium-induced autoimmune hemolysis was reported.1. This was followed by another report in 1995.2 The latter patient was a 24-month-old boy with sickle cell disease who had cardiac arrest and died 36 hours later from multiple organ failure. Other patients with serious nonfatal hemolysis from ceftriaxone have been described.34 In most (8/11) reported cases, the patients had a history of having received the drug.5

In a recent review of 11 cases of ceftriaxone-mediated autoimmune hemolytic anemia, the direct antiglobulin test was positive with anti-C3d, anti-immunoglobulin (Ig) G, or both in most cases.5 Ceftriaxone-dependent antibodies were detected in the serum by the indirect antiglobulin test in the presence of the drug, but not in its absence. These observations suggest binding of the drug-antibody complex to the red blood cells. In 1 of 11 cases, the antibody was reactive to a metabolite excreted in the patient's urine rather than to the native drug.5 Treatment of cells by the enzyme ficin was useful in cases in which the causative antibodies were hemolytic in nature. In these experiments, the patient's serum in the presence of the drug and a fresh source of complement demonstrated hemolysis of ficin-treated cells.

Serologie tests on serum or eluates, when performed in the absence of the drug, yielded negative reactions in the indirect antiglobulin tests. The serologie tests that included mixing of the serum or the eluate with the drug helped in differentiating drug-induced causes from other causes of immune hemolysis.

We recently observed severe hemolysis in a patient with sickle cell disease who was receiving ceftriaxone for pneumonia. This was the patient's second exposure to the drug. Using routine serologie studies, we demonstrated the binding of ceftriaxone antibody in the patient's serum by the immune complex mechanism.

Alterations in red cell autofluorescence were helpful in measuring erythrocyte response to oxidant stress induced by treatment of cells with hydrogen peroxide.6 In other experiments, umbilical cord erythrocytes have shown higher fluorescence compared with adult red cells.6 Furthermore, oxidant stress-related changes in autofluorescence have been correlated with lipid peroxidation and cell age in thalassemic red cells.7 Erythrocyte autofluorescence has also served as a marker for oxidative stress in red cells from patients with uremia.8 Based on these previous observations, we used flow cytometry to measure the changes in basal red cell autofluorescence induced by binding of the drug-antibody complexes to the red blood cells. The results of these studies are presented herein.

REPORT OF A CASE

A 10-year-old African American boy with sickle cell disease was admitted for pneumonia and was treated with intravenous ceftriaxone once daily. On admission, a blood transfusion was ordered because of anemia (hematocrit, 23%) in the presence of pneumonia. Routine serologie tests revealed that the patient had type A and Rh D-positive blood. An autocontrol was negative, and the serum showed no unexpected antibodies. A previously identified anti-C was not detected on this admission. The patient was transfused with 400 mL of C-negative, crossmatch-compatible red blood cells. No immediate posttransfuson reactions were reported. The patient was stable for the next 48 hours.

On the third hospital day, the patient had a sudden onset of generalized seizures and loss of consciousness, and he was found to have a hematocrit of 10%. Clinically, the patient did not show any bleeding, hemoglobinemia, or hemoglobinuria. Because of a marked decrease in the hematocrit value, the patient was transfused with 575 mL of C-negative red blood cells. Ceftriaxone was discontinued, and antibiotic therapy was switched to vancomycin hydrochloride and azithromycin. His hematocrit increased to 24% after the transfusion. Lactate dehydrogenase and serum haptoglobin levels were not measured. The patient required mechanical ventilation for less than 24 hours, promptly improved, and was discharged on the seventh hospital day with a hematocrit of 28%.

Hemolytic transfusion reaction to ceftriaxone was suspected because of the marked drop in hematocrit value in the absence of bleeding. Laboratory studies were performed to identify the suspected drug-induced reaction.

MATERIALS AND METHODS

Blood group and Rh typing, antibody screening, acid eluate preparation, ficin treatment of red blood cells, and dithiothreitol treatment of the serum were performed using routine blood bank techniques. Ceftriaxone was diluted with phosphate-buffered saline (pH 7.2) to achieve a concentration of 1 mg/mL.9 The diluted drug had a pH of 6.2 and was used in all subsequent serologie studies.

Testing for drug-dependent antibody was carried out with group O, Rh D-negative, C-negative cells treated with the drug solution for 1 hour at 37°C. The cells were washed and tested at 37°C with the patient's serum and the eluate from the patient's cells using a polyspecific antiglobulin reagent. A control specimen consisted of testing the patient's serum against untreated cells.

For studies to show immune complex formation, the patient's serum was incubated with the drug, and the mixture was subsequently tested against untreated cells. For these studies, an anti-IgG antiglobulin reagent was used. The controls consisted of normal serum or phosphate-buffered saline tested in a similar manner.

Flow cytometry studies were performed using a FAScan cytometer (Becton Dickinson, Immunochemistry Systems, San Jose, Calif). This flow cytometer is equipped with an argon laser with an excitation wavelength of 480 nm and an emission wavelength of 525 nm. Data analysis was performed using CellQuest software supplied by the manufacturer. A gate control, or threshold, was set for red cell autofluorescence on dot plot distributions of light scatter (forward and 90° side scatter) obtained with a 5% suspension of red blood cells from a blood group O, Rh D-negative, C-negative donor. Light scatter data of 10000 cells were collected and used to establish the threshold. The flow cytometer was used to test the serum or the phosphate-buffered saline incubated with the drug for 60 minutes at 37°C. Serial dilutions of the patient's serum were tested in a like manner.

RESULTS

A sample obtained at the time of reaction (3 days after the first transfusion) was retyped as group A, Rh D-positive, with no unexpected antibodies on screening. The direct antiglobulin test was positive (3+) with polyspecific antiglobulin reagent and with anti-C3d (1+).

Serologie studies to determine a ceftriaxone-dependent antibody with the posttransfusion sample and the eluate gave negative reactions with ceftriaxone-treated cells and untreated cells. A mixture of the patient's serum plus the drug, however, gave 4+ reactions at 37°C and 2+ reactions following the addition of antihuman globulin reagent. Normal serum plus the drug and the patient's serum plus buffer were tested simultaneously and gave negative results. The patient's serum, treated with 0.01 M dithiothreitol in the presence of the drug, gave negative reactions, confirming that the antibody in the serum was predominantly IgM. These studies were also carried out with ficintreated cells, and the results were unchanged compared with the non-enzyme-treated cells. In vitro hemolysis was not observed in any of the serologie studies, including those performed in the presence of the drug and those performed with ficin-treated cells.

Flow cytometry studies using normal red cells were used to establish the gating threshold for the basal red cell autofluorescence. When the red cells that had been previously incubated with the patient's serum plus the drug were tested in the flow cytometer, 28.81% of the cells were shown to be outside the threshold range. These data are shown in the Figure. Control studies also were performed using phosphate-buffered saline with the drug and showed an absence of fluorescence outside the established threshold. The patient's serum without the drug showed that only 0.96% of the cells manifested fluorescence outside the threshold (Table).

Dilution studies of the patient's serum also were carried out. The percentages of cells showing fluorescence outside the threshold with serial serum dilutions with fixed concentrations of the drug are shown in the Table.

COMMENT

Our patient had severe hemolysis based on the fact that, in the absence of any overt clinical bleeding, the patient's hematocrit decreased from 23% at the time of admission to 10%, when the patient suddenly deteriorated. The he molysis was associated with seizures and respiratory fail ure. More typical manifestations of hemolysis, such as he moglobinemia, hemoglobinuria, or jaundice, were not ob served. Our patient had received ceftriaxone 2 years pre viously. Prior exposure to ceftriaxone has been described in previously reported cases.5 Results of the serologie studies in our case suggest that IgM antibodies were in volved, based on the fact that the antibody activity was abolished by dithiothreitol treatment of the serum. Al though IgM antibodies have been shown to be hemolytic in vitro,1- we did not observe hemolysis during any serologie tests, including those performed with ficin -treated cells. Because the patient experienced seizures and respi ratory failure, it is possible that the IgM antibody could have caused red blood cell agglutination in vivo, which may have compromised perfusion of the vital organs, re sulting in these complications. Serious complications, such as cardiac arrest and even death, have been previously described in patients with ceftriaxone -induced acute he molysis.11,12 Further studies are needed to establish a cause -and -effect relationship between in vivo red cell ag glutination and the serious complications occurring in these patients.

Basal red blood cell autofluorescence has been demon strated previously by flow cytometry.6-8 This is due to the periodic acid -Schiff base compounds from aldehydes de rived from lipid peroxidation and amino groups of phos pholipids or cell proteins.6 The change in basal fluores cence has been demonstrated in patients with uremia and has been attributed to oxidative stress.8 In experiments that measured changes in autofluorescence, thalassemic red cells have shown an increased susceptibility to hydrogen peroxide -induced lipid peroxidation. In our patient, a change was detected from the basal autofluorescence for red blood cells after incubation with the patient's serum plus the drug. However, the mechanism responsible for the change is unclear. It is possible that the change may be due to the presence of ceftriaxone on the red blood cells and that ceftriaxone's fluorescence spectrum may be sim ilar to the excitation and emission spectra used in the flow cytometry experiments. However, we were unable to find information regarding the fluorescence spectrum of cef triaxone. Another possibility is that anticeftriaxone anti bodies bound to the red cells may be responsible for the change, perhaps owing to the oxidative stress caused by the antibody binding. Further studies are needed to un derstand the change in basal fluorescence observed when the drug -dependent antibody is bound to red blood cells.

In summary, our case confirms the previous serologie observations regarding ceftriaxone -induced hemolysis. In our case, flow cytometry results complemented the sero logie studies. Because flow cytometry is a sensitive and objective technique to analyze complex cell populations, our method may be useful to explore mechanisms for drug -dependent hemolysis.

References

1. Carratty C, Postoway N, Schwellenbach ), McMahill PC. A fatal case of ceftriaxone (Rocephin)-induced hemolytic anemia associated with intravascular immune hemolysis. Transfusion. 1991;31:176 -179.

2. Bernini JC, Mustafa MM, Sutor L], Buchanan GR. Clinical and laboratory observations: fatal hemolysis induced by ceftriaxone in a child with sickle cell anemia. J Pediatr. 1995;126(pt 11:813-815.

3. Longo F, Hastier P, Buckley MJ, Chichmanian RM, Delmont JP. Acute hep atitis, autoimmune hemolytic anemia, and erythroblastopenia induced by ceftri axone. Am I Castroenterol. 1998:93:836 -837.

4. Maraspin V, Lotric -Furlan S, Strle F. Ceftriaxone associated hemolysis. Wien Klin Wochenschr. 1999:111:368-370.

5. Seltsam A, Salama A. Ceftriaxone -induced immune haemolysis: two case reports and a concise review of the literature. Intensive Care Med2000;26:1390-1394.

6. Hammouda AMA, Fakeir A. Flow cytometric evaluation of erythrocyte re sponse to oxidant stress. Cytometry. 1995:20:19-22.

7. Senok AC, Li K, Nelson EAS, Arumanayagam M, Li CK. Flow cytometric assessment of oxidant stress in age -fractionated thalassemic trait erythrocytes and its relationship to in vitro growth of Plasmodium falciparum. Parasitology. 1998; 116(pt 11:1 -6.

8. Stoya G, Klemm A, Baumann E, et al. Determination of autofluorescence of red blood cells (RBCs) in uremie patients as a marker of oxidative damage. CUn Nephrol. 2002;58:198 -204.

9. Brecher ME. Detection of antibodies to penicillin or cephalosporins by test ing drug -treated red cells and demonstration of immune -complex formation in volving drugs. In: Brecher ME, ed. Technical Manual. 14th ed. Bethesda, Md: American Association of Blood Banks; 2002:722 -724.

10. Citak A, Garratty G, Ucsel R, Karabocuoglu M, Uzel N. Ceftriaxone -in duced haemolytic anaemia in a child with no immune deficiency or haemato logical disease. J Pediatr Child Health. 2002:38:209-210.

11. Borgna -Pignatti C, Bezzi TM, Reverberi R. Fatal ceftriaxone -induced he molysis in a child with acquired immunodeficiency syndrome. Pediatr Infect Dis J. 1995;14:1116-1117.

12. Viner Y, Hashkes PJ, Yakubova R, Segal -Kupershmit D, Luder AS. Severe hemolysis induced by ceftriaxone in a child with sickle -cell anemia. Pediatr Infect Dis J. 2000:19:83-85.

Ram Kakaiya, MD; Jill Cseri, SBB(ASCP); Steve Smith; Simone Silberman, MDt; Tara C. Rubinas, MD; Alan Hoffstadter, MT

Accepted for publication April 8, 2004.

From LifeSource Blood Services, Clenview, III (Dr Kakaiya, Ms Cseri, and Mr Smith); and Loyola University Medical Center, Maywood, III (Drs Silberman and Rubinas and Mr Hoffstadter).

The authors have no relevant financial interest in the products or companies described in this article.

[dagger] Deceased.

Corresponding author: Ram Kakaiya, MD, LifeSource Blood Services, 1205 N Milwaukee Ave, Clenview, IL 60025 (e-mail: rkakaiya@ itxm.org).

Reprints not available from the authors.

Copyright College of American Pathologists Aug 2004
Provided by ProQuest Information and Learning Company. All rights Reserved