tollefsen030907

Slide1 : Drug-induced Thrombocytopenia Doug Tollefsen March 9, 2007


Slide2 : 59 year-old woman admitted to BJH in 2004 with severe thrombocytopenia HPI: In generally good health until 1 day PTA, when she began to notice (1) petechial rash on the LEs, progressing to the trunk and arms; (2) bruising around the ankles; (3) bleeding after brushing her teeth; (4) blood-streaked stool. No previous history of bleeding or thrombocytopenia. Otherwise asymptomatic. PMH: Hypertension, hypercholesterolemia, osteoarthritis, GE reflux Allergies: None FH/SH/ROS: Non-contributory Meds: atorvastatin (Lipitor) begun 5 days PTA; previously on pravastatin celecoxib (Celebrex) esomeprazole (Nexium) hydrochlorothiazide and triamterene (Dyazide) niacin aspirin vitamin E calcium quinine PRN for leg cramps (took 1 day PTA)


Slide3 : Exam: Afebrile, BP 140/76, P 76 Obese, well appearing woman in no acute distress Petechiae/ecchymoses as per HPI No splenomegaly or lymphadenopathy Labs: Hgb 12.5, Hct 36.9 WBC 10,100 Platelets 4,000 (previously 169,000-208,000, last checked 2 yr PTA) No schistocytes or other abnormalities on peripheral smear Normal electrolytes, creatinine, glucose, LFTs Normal PT/aPTT, fibrinogen, D-dimer ANA positive 1:80 DDx: ITP vs. drug-induced thrombocytopenia Hospital course: Day #1: D/Ced Lipitor, Celebrex; continued Nexium, Diazide Given corticosteroids, IVIg, RhoGAM, and platelet transfusion Platelet count increased to 36,000 Day #2: No extension of purpura Platelet count increased to 66,000 Discharged on prednisone 100 mg/day Referred to hematology clinic for followup


Slide4 : Which of the latter drugs are most likely to cause thrombocytopenia? Case-control study Kaufman et al., Blood 1993; 82: 2714-2718 Individual case reports Reviewed in George et al., Ann Intern Med 1998; 129: 886-890 Updated Ann Intern Med 2001; 134: 346 Ann Intern Med 2003; 138: 239 Ann Intern Med 2005; 142: 474-475 Drug Safety 2007; 30: 185-186 What is the frequency of drug-induced thrombocytopenia? Estimates vary: ~1% of patients receiving GPIIb/IIIa antagonists ~1% of patients receiving unfractionated heparin <1:10,000 for most other drugs (causal relationship is often uncertain)


Slide5 : Blood 1993; 82: 2714-2718 Case-control study ~90 East Coast hospitals, 1983-1991 Identified 62 patients >16 yo admitted for (1) acute onset of bleeding (1st day of bleeding = “index day”) (2) platelets ≤30,000 (other counts normal) (3) rapid recovery of platelets to ≥150,000 without splenectomy or steroids (7 pts) with steroids rapidly tapered (55 pts) Median age 49, 65% female Exclusions: chemotherapy, radiotherapy, splenomegaly, DIC, HIV, SLE, mononucleosis, malignant blood disease, megaloblastic anemia, renal failure, cirrhosis, granulomatous disease, etc., etc. 2625 age/sex matched controls admitted for trauma (47%) acute infections (31%) other conditions (22%)


Slide6 : Blood 1993; 82: 2714-2718 (acetaminophen) Case-control study ~90 East Coast hospitals, 1983-1991 Identified 62 patients >16 yo admitted for (1) acute onset of bleeding (1st day of bleeding = “index day”) (2) platelets ≤30,000 (other counts normal) (3) rapid recovery of platelets to ≥150,000 without splenectomy or steroids (7 pts) with steroids rapidly tapered (55 pts) Median age 49, 65% female Exclusions: chemotherapy, radiotherapy, splenomegaly, DIC, HIV, SLE, mononucleosis, malignant blood disease, megaloblastic anemia, renal failure, cirrhosis, granulomatous disease, etc., etc. 2625 age/sex matched controls admitted for trauma (47%) acute infections (31%) other conditions (22%)


Slide7 : Blood 1993; 82: 2714-2718 (acetaminophen) Case-control study ~90 East Coast hospitals, 1983-1991 Identified 62 patients >16 yo admitted for (1) acute onset of bleeding (1st day of bleeding = “index day”) (2) platelets ≤30,000 (other counts normal) (3) rapid recovery of platelets to ≥150,000 without splenectomy or steroids (7 pts) with steroids rapidly tapered (55 pts) Median age 49, 65% female Exclusions: chemotherapy, radiotherapy, splenomegaly, DIC, HIV, SLE, mononucleosis, malignant blood disease, megaloblastic anemia, renal failure, cirrhosis, granulomatous disease, etc., etc. 2625 age/sex matched controls admitted for trauma (47%) acute infections (31%) other conditions (22%)


Slide8 : Criteria for review of case reports: 1. The candidate drug (a) preceded thrombocytopenia and (b) recovery from thrombocytopenia was complete and sustained after the drug was discontinued (unlikely if criterion 1 not met – Level IV) 2. The candidate drug was (a) the only drug used prior to the onset of thrombocytopenia or (b) other drugs were continued or reintroduced after discontinuation of the candidate drug with a sustained normal platelet count 3. Other etiologies for thrombocytopenia were excluded 4. Re-exposure to the candidate drug resulted in recurrent thrombocytopenia (strongest evidence)


Slide9 : Definite – Level I Probable – Level II Possible – Level III Criteria for review of case reports: 1. The candidate drug (a) preceded thrombocytopenia and (b) recovery from thrombocytopenia was complete and sustained after the drug was discontinued (unlikely if criterion 1 not met – Level IV) 2. The candidate drug was (a) the only drug used prior to the onset of thrombocytopenia or (b) other drugs were continued or reintroduced after discontinuation of the candidate drug with a sustained normal platelet count 3. Other etiologies for thrombocytopenia were excluded 4. Re-exposure to the candidate drug resulted in recurrent thrombocytopenia (strongest evidence)


Slide10 : Exclusion criteria: 1. Insufficient clinical data 2. Platelet count ≥100,000/µL 3. Use of a myelosuppressive agent 4. Nontherapeutic agents (e.g., environmental toxins, illicit drugs, drug overdose, obsolete drugs) 5. Drug-induced disease that included thrombocytopenia but predominantly involved other abnormalities (e.g., aplastic anemia, TTP) 6. Patient age ≤16 years


Slide11 : Included in the review: 515 patient case reports 48% of which supported “definite” (Level I) or “probable” (Level II) association between a drug and thrombocytopenia For Level I drugs, time to onset of thrombocytopenia: median 14 days (range 1-1000 days) time to recovery after stopping the drug: median 7 days (range 1-60 days) time to platelet nadir after re-challenge: median 3 days (range <1-60 days) time to recovery after re-challenge: median 5 days (range <1-60 days)


Slide12 : http://moon.ouhsc.edu/jgeorge Drug-induced Thrombocytopenia Level I 2001 Indinavir (Crixivan) (2 case reports) Atorvastatin (Lipitor) (1 case report) Pentoxifylline (Trental) (1 case report) Mesalamine (Asacol, Pentasa, Rowasa) (1 case report) 2003 Octreotide (Sandostatin) (1 case report) Abciximab (ReoPro) (group data) Eptifibatide (Integrilin) (group data) 2005 Rituximab (Rituxan) (1 case report) Tirofiban (Aggrastat) (group data) 2007 Adefovir dipivoxil (Preveon, Hepsera) (1 case report) Lopinavir/ritonavir (Kaletra) (1 case report) Teicoplanin (Targocid) (1 case report) Level II 2001 Ticlopidine (Ticlid) (2 case reports) Acetazolamide (Diamox) (2 case reports) 2003 Lotrafiban (5 case reports) Naproxen (Aleve and others) (3 case reports) Sulfamethoxypyridine (3 case reports) Chlorpropamide (Diabinese) (2 case reports) Roxifiban (2 case reports) Sulfapyridine (2 case reports) 2005 Chlordiazepoxide-clidinium (Librax) (2 case reports) Clopidogrel (Plavix) (2 case reports) Terbinafine (Lamisil) (2 case reports) Simvastatin (Zocor) (2 case reports) 2007 Efalizumab (Raptiva) (5 case reports) Etretinate (Tegison) (2 case reports) Oxaliplatin (Eloxatin) (2 case reports) Famotidine (Pepcid) (group data) (2 or more case reports) (1 or more case reports) (died) (9%) (28%) our pt D/Ced our pt cont’d our pt D/Ced


Slide13 : http://moon.ouhsc.edu/jgeorge Drug-induced Thrombocytopenia Level I 2001 Indinavir (Crixivan) (2 case reports) Atorvastatin (Lipitor) (1 case report) Pentoxifylline (Trental) (1 case report) Mesalamine (Asacol, Pentasa, Rowasa) (1 case report) 2003 Octreotide (Sandostatin) (1 case report) Abciximab (ReoPro) (group data) Eptifibatide (Integrilin) (group data) 2005 Rituximab (Rituxan) (1 case report) Tirofiban (Aggrastat) (group data) 2007 Adefovir dipivoxil (Preveon, Hepsera) (1 case report) Lopinavir/ritonavir (Kaletra) (1 case report) Teicoplanin (Targocid) (1 case report) Level II 2001 Ticlopidine (Ticlid) (2 case reports) Acetazolamide (Diamox) (2 case reports) 2003 Lotrafiban (5 case reports) Naproxen (Aleve and others) (3 case reports) Sulfamethoxypyridine (3 case reports) Chlorpropamide (Diabinese) (2 case reports) Roxifiban (2 case reports) Sulfapyridine (2 case reports) 2005 Chlordiazepoxide-clidinium (Librax) (2 case reports) Clopidogrel (Plavix) (2 case reports) Terbinafine (Lamisil) (2 case reports) Simvastatin (Zocor) (2 case reports) 2007 Efalizumab (Raptiva) (5 case reports) Etretinate (Tegison) (2 case reports) Oxaliplatin (Eloxatin) (2 case reports) Famotidine (Pepcid) (group data) (2 or more case reports) (1 or more case reports) (died) (9%) (28%) our pt D/Ced our pt cont’d our pt D/Ced


Slide14 : Some common drugs with uncertain relationship to thrombocytopenia ceftazidime (level III) ceftriaxone (level III) ciprofloxacin (level IV) furosemide (level IV) spironolactone (level III) valproic acid (level IV) dipyridamole (no case reports)


Slide15 : Proposed mechanisms of drug-induced thrombocytopenia Warkentin, NEJM 2007; 356: 891 GPIIb/IIIa or GPIb/IX/V “Quinine-type”


Slide16 : Characteristics of “quinine-type” anti-platelet antibodies Usually IgG and/or IgM React with GPIIb/IIIa and/or GPIb/IX/V only when the drug is present in soluble form Do not react with normal platelets pre-incubated with drug and then washed Do not react with washed platelets isolated from patients taking the drug Antibody binding to platelets is strongest when the drug is present at high concentration No inhibition of binding when drug is present in great excess over antibody Therefore, not typical “hapten-like” binding of drug to antibody Binding of drug to GP most likely produces a “compound” epitope or induces a conformational change elsewhere in the GP to which the antibody binds Aster, Toxicology 2005; 209: 149


Slide17 : Studied 3 quinine-dependent antibodies that react with human (not rat) GPIIIa Tested antibody binding to human/rat chimeric proteins to localize the binding site


Slide18 : Log Fluorescence Number of Cells Typical histogram obtained with sulfamethoxazole-dependent antibodies. Serum (60 µl) was incubated with 2.5 x 107 normal platelets ± 1.2 mM SMX. After three washes in buffer containing SMX at the same concentration, platelet-bound IgG was detected with fluorescein-labeled anti-IgG. Pt serum + SMX Pt serum - SMX Nl serum + SMX FACS assay for drug-dependent anti-platelet antibodies Curtis et al., Blood 1994; 84: 176


Slide21 : Vancomycin-dependent IgG and/or IgM were detected in 34 samples referred to Blood Center of Wisconsin (2001-2005); clinical information (e.g., platelet counts) available for 29 ~20% of samples referred for vancomycin testing No vancomycin-dependent antibodies were detected in 25 patients given vancomycin who did not develop thrombocytopenia 10 patients with quinine-induced thrombocytopenia 451 normal controls Two patients with severe thrombocytopenia temporally related to vancomycin (no other apparent cause) had negative antibody testing


Slide22 : Thrombocytopenic patients with vancomycin-dependent antibodies mean age 66 (range 37-87), 15 male/14 female mean nadir = 13,600 platelets 34% had “wet purpura”


Slide23 : Back to our patient . . . Hematology followup (2 weeks after discharge): Platelet count 147,000 (increased from 66,000 at discharge, nadir of 4,000) Prednisone taper begun Serum sent to Blood Center of Wisconsin for drug-dependent anti- platelet antibody tests: atorvastatin (Lipitor) quinine


Slide24 : Back to our patient . . . Hematology followup (2 weeks after discharge): Platelet count 147,000 (increased from 66,000 at discharge, nadir of 4,000) Prednisone taper begun Serum sent to Blood Center of Wisconsin for drug-dependent anti- platelet antibody tests: atorvastatin (Lipitor) negative quinine positive Further followup: Patient advised to avoid quinine Normal platelet counts maintained after discontinuing prednisone and re-starting Lipitor