Article Category: Clinical vignette
Published Online: Jan 31, 2017
Page range: 81 - 85
DOI: https://doi.org/10.5372/1905-7415.1001.469
Keywords
© 2016 Chantana Polprasert, Arunrat Pirunsarn, Ponlapat Rojnuckarin, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
Drug-induced immune hemolysis (DIIHA) is a potential complication from drug administration and can be a serious condition. The estimated prevalence is approximately 1 in 1 million of the population [1]. Because the laboratory tests for this condition are not generally available, the true prevalence may be higher. The severity of this condition varies from mild to severe cases that need medical treatment and drug cessation. There are more than one hundred drugs which are listed to cause DIIHA, such as antibiotics and chemotherapeutics. Cephalosporins and piperacillin have been the leading culprits for this entity [1, 2].
Suggested mechanisms for DIIHA include drug-independent antibodies, drug-dependent antibodies and nonimmunologic protein adsorption. Some drug-dependent antibodies can be detected by testing drug-treated red blood cells (RBCs) (“drug-adsorption type”); other antibodies can only be detected by testing in the presence of the drug (“immune complex type”) [3, 4]. Regardless of the mechanism, patients with DIIHA typically present with a positive direct antiglobulin test (DAT).
Corticosteroid is an immunosuppressive agent and is very unlikely to cause an autoimmune phenomenon. Therefore, it has been rarely reported as a cause of DIIHA. Hydrocortisone has been shown to induce intracellular hemolysis [5] and hemagglutination [6]. To our knowledge, prednisolone has not yet been reported to cause DIIHA.
Physicians should be aware of DIIHAs and the possibility of prednisolone-induced complement-mediated immune hemolysis by an ‘immune complex’ mechanism, and request appropriate diagnostic tests as indicated. Immunohematologic tests may be helpful for confirmation of diagnosis and selection of alternative drugs.
A 59 year-old male patient presented with bulky axillary lymphadenopathy His lymph node pathology was compatible with diffuse large B-cell lymphoma; stage IAX, according to the Ann Arbor staging system. The patient had a medical history of hemoglobin H disease, but he had rarely received blood transfusion. His baseline hemoglobin level was 10 gm/dL, mean corpuscular volume (MCV) was 61 fL and he had mild splenomegaly. The patient was scheduled to receive the CHOP regimen for lymphoma treatment. The CHOP regimen consisted of cyclophosphamide 750 mg/m2, vincristine 1.4 mg/m2, doxorubicin 50 mg/m2 and prednisolone 100 mg for 5 days every 3 weeks.
Twenty days after his first course, he developed progressive fatigue during hospitalization. A physical examination revealed markedly pale conjunctivae and nonicteric sclera. A complete blood count showed hemoglobin 6.6 gm/dL, hematocrit 3%, MCV 70 fL, WBC 5.59 x 109/L, and platelets 231 x 109/L. Blood chemistry found the total bilirubin 0.4 mg/dL (normal range 0.1–1.1 mg/dL), direct bilirubin 0.1 mg/dL (normal range 0.1–0.5 mg/dL), lactate dehydrogenase 834 U/L (313-618 U/L). Peripheral blood smear showed multiple clumps of RBC agglutination as shown in
Figure 1
Clumps of red cell agglutination from blood smear of the patient. Bar approx. 10 μm.
The direct antiglobulin test result was positive 1+for C3d, but negative for IgG. An indirect antiglobulin test was negative. The patient was diagnosed as having autoimmune hemolytic anemia. After RBC transfusion of 1 unit was administered, the anemic symptoms gradually improved. Cold agglutinin titer was 128 (normal range < 32). Immunoglobulin levels were within normal range; IgG 1340 mg/dL (normal range 700-1600 mg/dL), IgM 217 mg/dL (normal range 40–230 mg/dL) and IgA 185 mg/dL (normal range 70–400 mg/dL). The next 5 courses of chemotherapy were continued, but prednisolone was replaced by dexamethasone. The patient did not develop immune hemolytic anemia after the drug replacement.
This case report and any associated retrospective review of patient records was approved by the Ethics Committee of Chulalongkorn university (IRB No. 146/ 59, approval No. 008/2016) and has been sufficiently anonymized so as not to cause harm to the patient or their family. The patient has provided written informed consent for the publication of their case.
To prepare drug-treated RBC, 400 mg of the drugs were dissolved in 10 mL of phosphate-buffered saline (PBS) and then 1 mL of group O red cells added [7]. In a separate tube, control cells were prepared by adding 1 mL of untreated red cells (without the drug) to 10 mL of PBS. Both tubes were incubated for 1 hour at 37°C with occasional mixing. Cells were washed three times and prepared as a 5% suspension in PBS. Two sets of tubes were labeled (drug-treated and untreated) for each sample to be tested: patient’s serum, patient’s eluate, PBS, and pooled plasma. We modified the American Association of Blood Banks Technical Manual method using a gel (LISS/Coombs gel card, screening I and II, Bio-Rad Laboratories, Diamed, Cressier, FR, Switzerland) instead of a tube test. Salama et al. described the gel method for drug antibody detection [8]. We added 25 μL of each sample to the appropriate tube and added 50 μL of RBC (drug-treated and untreated set) to the gels. Gels were incubated at 37°C for 60 min. The tubes were centrifuged and examined for hemolysis and agglutination.
Drugs were prepared as a 1 mg/mL solutions in PBS and then centrifuged to remove any particulate matter [7]. We prepared 2 sets (untreated and enzyme-treated) of 6 gels for the following test mixtures: (a) patient’s serum + drug, (b) patient’s serum + PBS. (c) patient’s serum + complement (normal serum) + drug, (d) patient’s serum + complement (normal serum) + PBS, (e) Normal serum + drug, and (f) normal serum + PBS. We used papain for enzyme-treated red cells. We added 50 μL of each component to the appropriate gel, followed by 50 μL of a 5% saline suspension of the untreated group O reagent red cells to one set of tubes, and 50 μL of a 5% saline suspension of enzyme-treated group O reagent red cells to the second set of tubes. The components were mixed and incubated at 37°C for 1 to 2 hours. The tubes were centrifuged and the reaction examined.
Complement mediated hemolysis is usually the main mechanism for the ‘immune complex’ formation type of DIIHA, but only a rare cause in the drug adsorption type. To confirm the specific drug that caused complement mediated immune hemolysis, we performed an indirect antiglobulin test for all of the drugs given simultaneously to the patient. They were cyclophosphamide (Baxter), vincristine (Pfizer), doxorubicin (Pharmacia), and prednisolone (generic, Patar Lab). Specific tests for detection of drug antibodies in DIIHA were performed using both techniques, which were testing drug-treated RBCs (drug absorption type) and antibody testing in the presence of drug (immune complex type) [7, 9].
We prepared the drug-treated RBCs for the drug absorption type test by incubating the drug with group O RBCs for 1 h. Subsequently, we performed the indirect antiglobulin test using a gel card technique. We used both the drug-treated and the untreated negative control tubes. Samples tested included the patient’s serum, patient’s eluate, PBS, and pooled plasma. All 4 drugs suspected of causing DIIHA gave negative results.
The drugs suspected of causing DIIHA were prepared as a 1 mg/mL solutions in PBS for the immune complex method. Pooled group O RBCs, in 5% suspension, were either incubated with a proteolytic enzyme or left untreated. Two sets of group O RBCs (enzyme-treated and untreated) were tested in 6 panels according to this method. An indirect antiglobulin test for 2 sets of group O RBCs using a gel-test was performed. Only prednisolone gave a positive result as shown in
Results of tests using the immune complex method with prednisolone, dexamethasone, cyclophosphamide, vincristine, and doxorubicin (and enzyme-treated red blood cells)
| Drug | Patient’s serum | Normal serum | ||||
|---|---|---|---|---|---|---|
| With drug | With PBS | With complement and drug | With complement and PBS | With drug | With PBS | |
| Prednisolone | negative | negative | positive (1+) | negative | negative | negative |
| Dexamethasone | negative | negative | negative | negative | negative | negative |
| Cyclophosphamide | negative | negative | negative | negative | negative | negative |
| Vincristine | negative | negative | negative | negative | negative | negative |
| Doxorubicin | negative | negative | negative | negative | negative | negative |
PBS, phosphate-buffered salineo
The results of the test using the immune complex method suggested that prednisolone caused complement-mediated immune hemolytic anemia. Hydrocortisone-associated hemagglutination has also been reported [6]. In this and a related case, IgM antibody was found in the serum, but it did not cause in vivo or in vitro hemolysis [6, 10]. Hydrocortisone-induced intravascular hemolysis was positive in both drug-treated RBC and immune complex methods [5].
Because the present patient did not show any signs and symptoms of respiratory tract infection, infectious associated cold agglutinin disease [11] e.g.,
Matinengo et al. reported that hydrocortisone-dependent antibodies also reacted with other corticosteroid products, including dexamethasone, methylprednisolone, and betamethasone [5]. Because our patient did not develop immune hemolysis after being switched to dexamethasone, the antibody in this case appeared to be specific to prednisolone. Dexamethasone showed negative results in the in vitro test suggesting a role of this drug antibody detection method in predicting clinical hemolysis.
The excipients in the prednisolone tablets, produced by Patar Lab pharmaceutical manufacturer (Bangkok, Thailand), are starch, lactose, erythrosine, and magnesium stearate. Magnesium stearate is widely used as the lubricant for compressed tablets, while erythrosine is also commonly used for the color of the tablets. Lactose was reported to cause hemaglutination in a healthy donor [12], but this donor did not develop symptomatic hemolytic anemia. Pure prednisolone was not tested, but these 4 excipients are unlikely to be a cause of immune hemolysis.
In the immune complex mechanism, the 3 proposed components are: (1) drug, (2) RBC membrane, and (3) antibody. The drug-dependent antibody has been reported to be sometimes specific to the subtype of RBC antigen such as Rh, Kell, or Kidd blood group [13, 14]. Unfortunately, we did not analyze the RBC subtype in the present patient. Because corticosteroid-induced DIIHA is rarely encountered, some host factors, especially specific RBC alloantigens, should be further explored. Autoimmune hemolytic anemia (AIHA) is rarely associated with lymphoma except in chronic lymphocytic leukemia (CLL) [15, 16]. The prevalence of non-Hodgkin’s lymphoma, excluding CLL, associated with AIHA is approximately 0.23%–2.6% [6, 17] Generally, AIHA occurs in non-Hodgkin’s lymphoma while in the active disease phase, but not after treatment. AIHA is reported to be a rare complication in Thalassemia including hemoglobin H disease [18]. Immune activation may be caused by sequestered RBCs in the spleen [19] and by red cell fragmentation [20]. AIHA associated with Thalassemia requires steroids for treatment, but in the present patient the hemoglobin level gradually improved after blood transfusion and prednisolone cessation. Following the clinical presentation of this patient, neither lymphoma nor hemoglobin H disease were considered to be the cause of AIHA.
We recommend clinicians be aware of possible prednisolone-induced immune hemolysis. Immunohematologic tests may be helpful to confirm diagnosis and selection of alternative drugs.