Reports of thrombocytopenia associated with complementary and alternative medicines, herbal remedies, nutritional supplements, foods and beverages


Reproduced from the Eur J Haematol. 2010; 84: 421-429 [Full Text]


For the criteria for article assessment and the definition of levels of evidence click here



Case report summary

Substances with definite evidence for a causal association with thrombocytopenia


33 M with ulcerative colitis developed thrombocytopenia (5 x 103/µl) and rectal bleeding requiring transfusions.  Platelet count recovered after food intake discontinued and remained normal until cow’s milk was added to diet; then thrombocytopenia recurred (150 →11 x 103/µl). Platelet counts remained normal and ulcerative colitis resolved with a milk-free diet.

Cranberry juice23

68 M developed thrombocytopenia (1 x 103/µl) with mucocutaneous bleeding and hematuria. Platelet counts recovered to normal (200 x 103/µl) with steroid treatment.  Dietary history suggested cranberry juice, taken for 10 days for urinary retention, was associated with the thrombocytopenia. Using indirect immunoflorescence, cranberry juice-dependent platelet-reactive IgG and IgM antibodies were detected using dialyzed juice of the same brand and batch. Challenge with cranberry juice not reported.

Jui herbal tea24

51 F developed thrombocytopenia (16 x 103/µl) with mucocutaneous bleeding associated with Jui herbal tea taken several days before her annual health examination. Challenge with Jui caused thrombocytopenia (305 →2 x 103/µl) within one day.

Jui herbal tea25

54 M with liver cirrhosis had 3 episodes of thrombocytopenia with platelet counts of zero and mucocutaneous bleeding. Association with Jui was noted. Platelet counts following recovery from each episode are not reported but are described as “his usual range”. Challenge with Jui caused the platelet count to rapidly decrease to zero in 6 hours and recovered to “pre-challenge levels” on day 6. Thrombocytopenia did not recur after stopping Jui.

Lupinus termis bean26

13 M developed thrombocytopenia on 5 occasions after Lupinus termis bean ingestion, 3 times as a supervised challenge with a single bean. At the last challenge, the platelet count decreased from 160 →10 x 103/µl in 2 hours. Recovery platelet counts not reported.

Tahini 27

28 F with multiple episodes of thrombocytopenia (lowest platelet count, 6 x 103/µl) with mucocutaneous bleeding. The patient associated thrombocytopenia with ingestion of tahini. Following challenge, the platelet count fell from 161 to 34 x 103/µl in one day and recovered (189 x 103/µl) in 9 days.

Substances with probable evidence for a causal association with thrombocytopenia


38 F (Case 2) developed thrombocytopenia (70 x 103/ µl) and leukopenia (2270/µl) in addition to ataxia and paresthesias 2 hours after drinking Bajiaolian infusion. Platelet count recovered to normal (227 x 103/µl) in 12 days, leukopenia in 25 days, but paresthesias persisted.  Criteria 1, 2, and 3 were met.

Cupressus funebris (Mourning cypress)29

62 F developed acute systemic illness (acute renal failure, liver toxicity, hemolytic anemia and thrombocytopenia [nadir, 2 x 103/ µl on day 4]) 2 hours after ingesting hot water extract from sliced wood (C. funebris). Her first ingestion 1 week previously had caused no symptoms. “Recovery from thrombocytopenia within days” is stated but no recovery platelet counts are reported. Increased serum creatinine persisted for >1 month. Criteria 1, 2, and 3 were met.

Milk, Potato30

26 F (Case 2) with repeated episodes of nausea, vomiting, and bloody diarrhea associated with thrombocytopenia; milk and potato were suspected etiologies.  Challenge with potato: platelet count decreased from 104 to 91 x 103/µl at 30 minutes. Challenge with milk: platelet count decreased from 56 to 46 x 103/µl at 30 minutes. Following elimination of milk and potatoes from her diet, platelet counts were 140 to 190 x 103/µl for the following 9 months.  Criteria 1, 2, and 3 were met; challenge criterion not met.

Vitamin A palmitate31

3 month-old M developed thrombocytopenia (30 x 103/µl) and anemia. Marrow aspirate demonstrated decreased megakaryocytes and erythroid dysplasia. He had received 62,000 U of vitamin A each day (41-times the recommended requirement) since 10 days of age because of misunderstood instructions.  Platelet count recovered to 220 x 103/µl 30 days after vitamin A stopped. Criteria 1, 2, and 3 were met.

Substances with possible evidence for a causal association with thrombocytopenia

Chromium picolinate33

33 F developed acute renal failure, liver toxicity, hemolytic anemia and thrombocytopenia (15 x 103/µl) after taking chromium picolinate tablets for 4-5 months for weight loss. Plasma chromium concentration was more than twice normal. Platelet count and other abnormalities returned to normal by day 26 after stopping chromium tablets. Criteria 1 met; insufficient data to assess criteria 2 and 3. 

Echinacea pallida34

32 M used Echinacea pallida for treatment of a respiratory infection for one week.  Paracetamol was also used. Two weeks after stopping Echinacea he developed thrombocytopenia (20 x 103/µl), microangiopathic hemolytic anemia, and seizures.  He recovered following one month of plasma exchange treatment. Criterion 1 met; insufficient data to assess criteria 2 and 3

Hypericum perforatum (St. John’s Wort)35

22 M was admitted to hospital with fever, pharyngitis, anemia (6 g/dl), agranulocytosis (absolute neutrophil count, 12/µl). He had been taking St. John’s Wort for 3 weeks and fluoxetine for 3 months for depression. Thrombocytopenia developed on day 3 (464 →83 x 103/µl). Marrow biopsy demonstrated necrosis. The patient died on day 8. Although platelet count did not recover, criterion 1 may have been met if the patient had survived.  Criteria 2 and 3 not met; a viral infection etiology as well as St. John’s Wort toxicity was considered.


Patient (age, gender not reported) developed thrombocytopenia (platelet count decreased from 168 to 83 x 103/µl) after 12 weeks of treatment with nicotinamide for hyperphosphatemia while on chronic hemodialysis. Platelet count recovered to 180 x 103/µl 2 weeks after nicotinamide discontinued.  Criteria 1 met; insufficient data to assess criteria 2 and 3. 

Complete Thymic Formula37

Patient (age, gender not reported) with hepatitis C developed thrombocytopenia (4 x 103/µl) following 5 months use of the dietary supplement, Complete Thymic Formula. The patient was also taking naproxen; both were discontinued and “the platelet count returned to baseline”. Criteria 1 met; insufficient data to assess criteria 2 and 3. 

Substances with an unlikely association with thrombocytopenia


3 patients (Cases 3-5) developed thrombocytopenia (20-56 x 103/µl) after drinking Bajiaolian infusions, but data on recovery from thrombocytopenia are not reported, therefore criterion 1 not met.  All 3 patients also had neurologic abnormalities.


54 F developed thrombocytopenia (16 x 103/µl) with mucocutaneous bleeding after 6 weeks of kelp tablets (that contained 1.3µg/g arsenic) and a multivitamin.  Marrow aspirate demonstrated normal megakaryocytes and dyserythropoiesis. Platelet count recovered to 60 x 103/µl with treatment with steroids and azathioprine. Complete recovery not reported; criterion 1 not met.


F, age 10-60 days, born with thrombocytopenia-absent radius syndrome had recurrent diarrhea, eosinophilia, and thrombocytopenia associated with 2 periods of cow’s milk feeding. Diarrhea and eosinophilia, but not thrombocytopenia, recurred with a third introduction of cow’s milk. Criteria 1 and 4 not met.


Infant M born to mother with ITP had platelet count of 11 x 103/µl at birth that persisted, in spite of steroid treatment for 11 weeks, when milk stopped because of diarrhea.  Platelet count increased from 11→150 x 103/µl after 6 days (3 days, no milk; 3 days half-strength milk). After 10 days on normal milk, platelet count 30 x 103/µl; 3 weeks after stopping milk, platelet count 200 x 103/µl.  Milk resumed at age 1 year when platelet count platelet count 300 x 103/µl; mother reported bruising after 2 attempts to resume milk; no platelet counts reported. The challenges with milk appeared to fulfill criterion 4 but the diagnosis was uncertain because of the mother’s ITP and the unspecified nature of the milk products.

Niacin (nicotinic acid, vitamin B3)41

51 M (Case 2) developed thrombocytopenia (92 x 103/µl) and leukopenia (2.1 x 103/µl) while taking 2.5 g/day nicotinic acid for about 3 years. Platelet counts increased after stopping nicotinic acid, but remained low (118 x 103/µl) 2 months later.  Criterion 1 not met.

Herbal “Ultimate Colon Cleanse Program TM42

51 F developed thrombocytopenia (24 x 103/µl), neutropenia (24 x 103/µl) and anemia (8.5 g/l) following 30 days of this product which contained 39 different herbal ingredients. Marrow aspirate demonstrated aplasia (<5% cellularity). No follow-up reported. Criterion 1 not met.  

Substances in reports that were excluded from analysis


Case 1: 10 F developed thrombocytopenia (1 x 103/µl) and hepatosplenomegaly that resolved when aspartame stopped. Two recurrences after ingesting aspartame “with remissions when abstaining from aspartame”, but no data reported. Case 2: 11 F developed thrombocytopenia after chewing aspartame gum; resolved when aspartame was stopped; recurred with aspartame candy. No data reported. Case 3: 61 M developed thrombocytopenia (54 x 103/µl) after drinking diet colas for two years; resolved when aspartame stopped; “retest trial resulted in an immediate exacerbation”. No data reported. Case 4: F (age not reported) developed thrombocytopenia (30 x 103/µl) with hypertension following pregnancy while drinking diet colas; resolved when aspartame stopped.  There is insufficient data in these small patient descriptions to assess criteria.

Chinese herbal medicine44

Patient (age, gender not reported) developed thrombocytopenia (3 x 103 platelets/µl) with mucocutaneous bleeding. “The patient had taken an unidentified Chinese herbal medicine.” No additional data reported.

Foods (cocoa, egg, potato, wheat)30

5 case reports excluded because of insufficient evidence to assess criteria (Case 3) or platelet counts not less than 100 x 103/µl (Cases 1, 4, 5, 6)

Garbhapal ras45

F (age no reported) presented with thrombocytopenia after repeated use of this Ayurvedic medicine that contained multiple metals and herbal preparations. No platelet count data reported.

Niacin (nicotinic acid, vitamin B3)46

32 M developed thrombocytopenia (118 x 103/µl) and liver toxicity after taking niacin and other vitamins for 10 years.  Niacin stopped; other vitamins continued; platelet count recovered to 193 x 103/µl in 11 days; serum transaminase levels also returned to normal.  Excluded because platelet count not <100 x 103/µl.


6 patients on hemodialysis with hyperphosphatemia treated with nicotinamide, 1000 mg/day; 5 developed thrombocytopenia (mean values: 188 ± 17 x 103/µl before therapy; 122 ± 41 x 103/µl within 3 months of beginning nicotinamide; 150 ± 9 x 103/µl 10 days after discontinuing nicotinamide.  Excluded because data are insufficient to meet criterion for platelet count <100 ± 9 x 103/µl.


67 M with hepatitis C developed thrombocytopenia (5 x 103/µl) with mucocutaneous bleeding. Thrombocytopenia improved with intravenous immunoglobulin and steroid treatment, but platelet counts are not reported. Exposure one year previously was also associated with thrombocytopenia (98 x 103/µl). Reported data are insufficient to assess Criterion 1. 

Traditional African herbal remedies49

5 patients reported who had “complete absence of platelets” on the peripheral blood film.   Excluded because of insufficient evidence to assess criteria.






Reference List


  22. Caffrey EA, Sladen GE, Isaacs PET, Clark KGA. Thrombocytopenia caused by cow's milk. Lancet 1981;2:316.

  23. Davies JK, Ahktar N, Ranasinge E. A juicy problem. Lancet 2001;358:2126.

  24. Azuno Y, Yaga K, Sasayama T, Kimoto K. Thrombocytopenia induced by Jui, a traditional Chinese herbal medicine. Lancet 1999;354:304-5.

  25. Ohmori T, Nishii K, Hagihara A, Takeda M, Sekido K. Acute thrombocytopenia induced by Jui, a traditional herbal medicine. J Thromb Haemost 2004;2:1479-80.

  26. Lavy R. Thrombocytopenic purpura due to Lupinus termis bean. J Allergy Clin Immunol 1964;35:386-8.

  27. Arnold J, Ouwehand WH, Smith G, Cohen H. A young woman with petechiae. Lancet 1998;352:618.

  28. Kao WF, Hung DZ, Tsai WJ, Lin KP, Deng JF. Podophyllotoxin intoxication: toxic effect of Bajiaolian in herbal therapeutics. Human Exp Toxicol 1992;11:480-7.

  29. Lee JJ, Chen HC. Flavonoid-induced acute neuropathy by Cupressus funebris Endl (Mourning Cypress). Am J Kidney Dis 2006;48:e81-e85.

  30. Squier TL, Madison FW. Thrombocytopenic purpura due to food allergy. J Allergy 1937;8:143-54.

  31. Perrotta S, Nobili B, Rossi F, Criscuolo M, Iolascon A, Pinto D, Passaro I, Cennamo L, Oliva A, Ragione F. Infant hypervitaminosis A causes severe anemia and thrombocytopenia: evidence of a retinol-dependent bone marrow cell growth inhibition. Blood 2002;99:2017-22.

  32. Kojouri K, Vesely SK, George JN. Quinine-associated thrombotic thrombocytopenic purpura-hemolytic uremic syndrome: frequency, clinical features, and long-term outcomes. Ann Int Med 2001;135:1047-51.

  33. Cerrulli J, Grabe DW, Gauthier I, Malone M, McGoldrick MD. Chromium picolinate toxicity. Annals Pharmacotherapy 1998;32:428-31.

  34. George L, Ioannis E, Radostina T, Antonios M. Severe thrombotic thrombocytopenic purpura (TTP) induced or exacerbated by the immunostimulatory herb Echinacea. Amer J Hematol 2006;81:224.

  35. Demiroglu YZ, Yeter TT, Boga C, Ozdogu H, Kizilkilic E, Bal N, Tuncer I, Arslan H. Bone marrow necrosis: a rare complication of herbal treatment with Hypericum perforatum (St. John's Wort). Acta medica 2005;48:91-4.

  36. Takahashi Y, Tanaka A, Nakamura T, Fukuwatari T, Shibata K, Shimada N, Ebihara I, Koide H. Nicotinamide supresses hyperphosphatemia in hemodialysis patients. Kidney Internat 2004;65:1099-104.

  37. Raymond RS, Fallon MB, Abrams GA. Oral thymic extract for chronic hepatitis C in patients previously treated with interferon. Ann Int Med 1998;129:797-800.

  38. Pye KG, Kelsey SM, House IM, Newland AC. Severe dyserythropoiesis and autoimmune thrombocytopenia associated with ingestion of kelp supplements. Lancet 1992;339:1540.

  39. Whitfield MF, Barr DG. Cows' milk allergy in the syndrome of thrombocytopenia with absent radius. Arch Dis Child 1976;51:337-43.

  40. Jones RH. Congenital thrombocytopenia and milk allergy. Arch Dis Child 1977;52:744-5.

  41. O'Brien T, Silverberg JD, Nguyen TT. Nicotinic acid-induced toxicity associated with cytopenia and decreased levels of thyroxine-binding globulin. Mayo Clin Proc 1992;67:465-8.

  42. Smereck J. Aplastic anemia: a possible toxic effect of an herbal "colon cleansing" preparation. JEmergency Med 2009;36:191-3.

  43. Roberts HJ. Aspartame-induced thrombocytopenia. Southern Med J 2007;100:543.

  44. Taylor DM, Walsham N, Taylor SE, Wong L. Use and toxicity of complementary and alternative medicines among emergency department patients. Emergency Medicine Australasia 2004;16:400-6.

  45. Parab S, Kulkarni R, Thatte U. Heavy metals in 'herbal' medicines. Indian Journal of Gastroenterology 2003;22:111-2.

  46. Reimund E, Ramos A. Niacin-induced hepatitis and thrombocytopenia after 10 years of niacin use. J Clin Gastroenterol 1994;18:270-1.

  47. Rottemburg JB, Launay-Vacher V, Massard J. Thrombocytopenia induced by nicotinamide in hemodialysis patients. Kidney Internat 2005;68:2911-2.

  48. Kiguchi T, Kimura F, Kiiya K, Katayama Y, Harada M. Acute thrombocytopenic purpura after ingestion of Sho-saiko-to for hepatitis. Liver 2000;20:491.

  49. Lowenthal MN, Jones IG, Desai M. Aplastic anaemia and optic fundus haemorrhages due to traditional herbal remedies. J Tropical Med Hygiene 1978;81:177-9.