A Diagnostic Improvement Analysis: A-50-year old man with abdominal pain and thrombocytopenia.

The main objectives with the description and the analysis of this case is to show possible biases in the diagnostic process under the opinion of the analyst, searching for human cognitive and system aspects that could be improved. We do not know the real situation under which the attending physician had to act during the care of this patient

A Diagnostic Improvement Analysis: A-50-year old man with abdominal pain and thrombocytopenia

A-50-year old man was admitted to the hospital after several visits to the Emergency Department (ED) in a two month period. He was born in Nigeria, but he was living in Spain in the last 15 years, and the last trip to his natal country was 6 years ago. He was married with a heterosexual couple. He was a light smoker with hypertension and hyperuricemia. He tested positive for HIV, but he denied any  exposure.

He went to the ED referring abdominal pain located in the epigastrium and irradiated to both loin, with nausea and hyporexia and no fever. The symptoms started around six months before, but occasionally they  worsened.

Physical examination

Blood pressure 180/120 mmHg. Temperature: 35,7 ºC. Examination of cardiovascular, respiratory, abdominal and neurological systems is normal. No information about the skin.

Lab/ Radiology

Blood test: Hemoglobin 9,1 g/dl; Platelets count: 17000; WBC: 5310. Reticulocytes: 66%. Protrombin time: 81%.

Haptoglobin: 13 mg/dL (30-200 mg/dL)

Coombs test: positive.

Haemoglobin S test: 36,5%.

Peripheral smear extension: Anisopoikilocytosis with schystocytes (2,4%) . Some isolated sickle cells. Thrombocytopenia. No parasites.


Bilirubin: 1,2 mg/dL

Creatinine:2,2 mg/dL.

Glomerular filtration: 33 mL/min.

Sodium and Potassium normal.

Alanine transaminase: 57 IU/L. LDH:841

D-Dimer: 13016

Autoimmunity tests: negative.

Urinalysis: ++ blood; +++ protein

Blood cultures: negative

Abdominal sonogram: normal

Transthoracic Ecocardiogram: normal

Clinical evolution

Several diagnostic hypotheses were established before in each consultation to the ED and the patient received different schedule and drugs to treat the abdominal pain. The platelet count was going down progressively. In  the last contact with the  ED, the patient stayed at the hospital with the following preliminary diagnoses:

  1. Nephritic syndrome with renal insufficiency.

2 .Normocytic anemia and hemolysis, to discard malaria or thrombotic  thrombocytopenic purpura (TTP).

Final Diagnosis

Thrombotic Thrombocytopenic Purpura , based on the symptoms(abdominal pain probably associated to microthrombus in the vasculature), with negative blood cultures, and the presence of thrombocytopenia and data of  microangiopathic hemolytic anemia (1).

The patient was treated with plasmapheresis and steroids, with a total resolution of the clinical problem.

In a dermatological examination before starting the treatment, some petechiae were detected in the skin, and they were clearly visible in the palms and the soles.

There are recent Guidelines for the diagnosis of TTP (2) .


In summary, this is the history of a middle-age African born male, with a positive HIV test, mild hypertension, starting with centroabdominal pain and nausea, without fever. The lab test showed a progressive thrombocytopenia and a hemolytic anemia. The final diagnosis was TPT, possibly associated with HIV infection.

Diagnostic Improvement Analysis

Type of Error: Diagnosis delay. The patient went to the ED  five times during a two month period, before a definitive diagnosis.

Cognitive component:

  • Uncommon diagnosis: TTP is an emergency, but it is a rare disease, with signs and symptoms common to many other diseases. This is one of the problems that you have to consider always in a “worst-case scenario” in a patient with thrombocytopenia, abdominal pain and neurologic symptoms.
  • Representativeness heuristic: it could have happened that in the mind of the attending in the ED, a  patient with hemolysis, abdominal pain and of African origin fit a model of “malaria”, although the patient was afebrile.  Sickle cells are red-cell with a particular shape, due to the presence of Haemoglobin S, a genetic trait “protective” for malaria.
  • Data Gathering: Although the patient was of African origin, he lived in Spain for several years, making malaria unlikely. Furthermore, the physical examination does not describe wether there was presence of petechiae or not.
  • Premature closure: The doctor did not expand the diagnostic possibilities, and without a definitive verification, selected two diagnoses, the first based on hypertension and the presence of blood in the urine, and the second descriptive of the anemia.

System aspects

. In this case, with a “zebra” diagnosis,  the consultation with an “Electronic Diagnostic Decision  Tool” (EDDT) could have listed these rare diagnoses, such as the Thrombotic Thrombocytopenic Purpura.

Improvement measures
  • A good and accurate data collection reduces the probability of representativeness heuristic.
  • When the patient belongs to a different social community, the signs and symptoms may present differently or the physical examination may have to be adapted.
  • Try to make a complete  differential diagnosis, and don’t stop thinking when you find any combination of signs or symptoms.
  • Keep the method on physical examination, even in a summary form.
  • Use an EDDT to remember infrequent entities that cannot be reached by memory.
  1. Thrombotic thrombocytopenic purpura.  C.Tellez-Hinojosa, A.Vazquez-Mellado, D.Gómez-Almaguer.
    http://dx.doi.org/10.1016/j.rmu.2015.10.0011665-5796/©2016UniversidadAutonoma de Nuevo Leon.
  2. ISTH guidelines for the diagnosis of thrombotic
    thrombocytopenic purpura. J Thromb Haemost. 2020;18:2486–2495.  DOI: 10.1111/jth.15006

Author: Dr. Lorenzo Alonso