A late diagnosis of essential thrombocythemia in a patient with esophageal varices and splenomegaly due to portal vein thrombosis: a case report

¹Department of Internal Medicine, Baptist Institute of Health Sciences, Mbingo Baptist Hospital, Baingo, Cameroon, ²Department of Imaging, Mbingo Baptist Hospital, Baingo, Cameroon

^&Corresponding author
Lambi Jervis Somo, Department of Internal Medicine, Baptist Institute of Health Sciences, Mbingo Baptist Hospital, Baingo, Cameroon

Introduction    

Essential thrombocythemia (ET) is one of the chronic myeloproliferative neoplasms (MPNs), characterized by marked thrombocytosis, thrombotic and hemorrhagic risks, and constitutional symptoms. The incidence varies from 0.2 to 2.5 per 100,000 people per year, with a prevalence of 38 to 57 cases per 100,000 people, and the median age at diagnosis is 60 years [1]. The disease has a slight female preponderance [2]. Essential thrombocythemia clinically manifests as the overproduction of platelets, which is caused by mutations in JAK2 V617F, Calreticulin (CALR), or Myeloproliferative Leukemia (MPL) gene [3]. This leads to a high risk of thrombosis within arterial and venous territories [3], which includes the portal vein, and this can result in portal hypertension. Portal hypertension due to PVT can manifest as esophageal varices [4]. One of the main goals of management for patients with ET is to prevent bleeding and thrombosis [2,3]. However, this can be quite challenging given the diverse presentations, where preventing thrombosis may increase the risk of bleeding, as seen in this case. With limited resources in our setting, early diagnosis can be challenging. We describe the case of an adult female with a late diagnosis of ET after presenting with persistent epigastric pain and endoscopic findings of esophageal varices due to portal vein thrombosis.

Patient and observation     

Patient information: a 46-year-old Cameroonian female presented with complaints of left upper quadrant pain, which had been going on for three years at the time of presentation. The pain was gradual in onset, constant, dull, non-radiating, and mild to moderate in intensity with no known exacerbating or relieving factors. She also complained of having intermittent burning-like sensations at the tips of her fingers during this time. There were no changes in her vision. There was no pruritus, including on exposure to water. A review of other systems was generally unremarkable. She had no chronic illnesses or relevant family history. Her records revealed that she had been treated for antigen-positive malaria in the past. A complete blood count (CBC) from her medical records showed a white blood cell count (WBC) of 10.3 x 10⁹/L, Hemoglobin (Hb) of 11.2 g/dl, mean corpuscular Volume (MCV) of 72 fl, and platelet count of 1,086 x 10⁹/L. A peripheral blood smear (PBS) showed 3% ovalocytes and 1% teardrop cells. An abdominal ultrasound revealed a homogenous splenomegaly. She was initially diagnosed with hyper-reactive malarial splenomegaly (HMS). She was prescribed mefloquine 250mg, one tablet weekly, which she took for six months. A repeat abdominal ultrasound was done, which revealed persistent splenomegaly. She also received iron supplementation. On later review, there were no changes in her spleen size, the platelet counts remained elevated on CBC, and the pain persisted. Another PBS was done, revealing thrombocytosis. An internal medicine consult was requested, and she was referred to our facility.

Clinical findings: physical exam was notable for an enlarged, non-tender spleen with a regular outline. The spleen was palpated about 4cm below the left costal margin. There were no palpable peripheral lymph nodes. No hepatomegaly. The rest of the physical exam was unremarkable.

Timeline of current episode: the initial episode of pain was in March 2018. Treatment for antigen-positive malaria was initiated in January 2019. Initial CBC, PBS, and abdominal ultrasound in her records were done in April 2019. Treatment for HMS was between April and September 2019. A repeat abdominal ultrasound was performed in January 2020. A repeat PBS and CBC, revealing persistent thrombocythemia, was done in September 2020, and she was referred to an internist. An internist saw the patient in March 2021.

Diagnostic assessment: she underwent a bone marrow aspirate and biopsy. The results were non-diagnostic due to a lack of sufficient marrow elements. Two subsequent bone marrow biopsies, a few days later, revealed the same findings. Platelet counts remained elevated at 1,266 x 10⁹/L (Table 1). Abdominal Ultrasound revealed an enlarged spleen of 18.93cm (Figure 1). In addition to her high platelet counts, MCV at the time was 65, suggesting an underlying coexisting iron deficiency anemia, which could be a cause of secondary thrombocytosis. Essential thrombocythemia was suspected. In addition to ET, the differential diagnoses included secondary thrombocytosis due to iron deficiency anemia, primary myelofibrosis, and HMS.

Therapeutic interventions: despite non-diagnostic results from bone marrow biopsies, the patient was managed as having essential thrombocythemia. The diagnosis was not confirmed initially with JAK2 testing. She was considered low-risk. This was based on the persistently high platelet counts. At this point, it was unknown where JAK2 testing could be done in Cameroon. She received oral aspirin, 81mg per day, for three months. She also received iron supplementation.

Follow-up and outcome: the abdominal pain got worse, and the paresthesias of the fingertips persisted. There was a concern about the gastrointestinal side effects of aspirin, so a fecal occult blood test was requested, which returned positive. An upper GI endoscopy revealed Grade 2 lower esophageal varices and Gastric Varices. There were no stigmata of liver cirrhosis, and the liver function tests were normal. There was no evidence of cirrhosis on abdominal sonography. To investigate the source of the varices, Doppler studies of intra-abdominal vessels, including those of the portal venous system, were requested. The portal vein was occluded by a thrombus (Figure 2). In the context of Portal Vein (PV) thrombosis, she was categorized as a high-risk ET patient. We therefore commenced oral hydroxyurea 500mg BID and oral rivaroxaban 15mg BID. Platelet counts were at 886 x 10⁹/L at the time hydroxyurea was started (Table 1). Anticoagulation for PV thrombosis was later halted due to the risk of bleeding from Esophageal varices and also the possibility of acquired Von Willebrand disease, which is not uncommon in ET. Subsequent Doppler ultrasonography, about three months later, revealed extensive thrombosis involving the PV, splenic vein, and superior mesenteric vein (SMV) (Figure 3). She also developed cavernous transformation of the portal vein (Figure 4), and splenic vein collaterals through the renal Vein (Figure 5).

Eventually, she was sent to test for JAK2 V617F mutation, BCR-ABL translocation, and CALR. The JAK2 V617F test was positive with 38.2% mutation in the allele. BCR-ABL and CALR were negative. This confirmed the diagnosis of high-risk ET with PV thrombosis and esophageal varices. A follow-up endoscopy was done, revealing progression of the esophageal varices from grade 2 to 3 with gastric varices present. Three varices were banded. A consult was sent to a hematologist, and the decision was made to continue with Hydroxyurea to target platelet counts between 100 x 10⁹/L and 400 x 10⁹/L. The doses of Hydroxyurea were adjusted accordingly to achieve this target. CBC was usually done monthly (Table 1). Anticoagulation was restarted with oral rivaroxaban at 20mg QD. Aspirin was discontinued due to the high risk of bleeding when combined with rivaroxaban. She also received a beta blocker, oral Propranolol 20mg BID for variceal hemorrhage prophylaxis. For the abdominal pains, she has been receiving paracetamol 1g TID as needed, which has been fairly effective in controlling the pain, and omeprazole 20mg daily orally. She continued to receive iron supplementation, and her MCV rose to 79 from an initial value of 65. In addition, she is receiving folic acid. Timeline and changes in CBC parameters while on hydroxyurea are recorded in Table 1; Hydroxyurea was initiated on 23/2/22; not all the follow-up values are included here. Platelet counts were normalized. To date, she has not reported any episodes of hematemesis. She reports improvement in the abdominal pain. Follow-up endoscopy will be done every six months to one year.

Patient perspective: “I am very pleased that my issue was properly diagnosed after consulting for many years. If my case is published online, people will have more awareness about this condition.”

Informed consent: the patient provided informed consent for the publication of this medical case in a medical article.

Discussion     

We report the case of a middle-aged Cameroonian woman who presented with persistently elevated platelet counts and left upper quadrant pain, with splenomegaly. The patient was eventually diagnosed with Essential Thrombocythemia. She was 46 years old at the time of diagnosis. Her gender correlates with the slight female-to-male predominance of ET [2]. Despite her low MCVs, which could point towards a diagnosis of secondary thrombocytosis due to iron deficiency anemia, it is not improbable for both ET and Iron deficiency anemia to co-exist. This was demonstrated in a case report by Namdaroglu et al. [5] in 2016 of a 56-year-old male who had concomitant Iron deficiency anemia, ET, and Renal cell carcinoma. Also noteworthy is that our patient received iron supplementation, and her MCV returned to near normal, but thrombocytosis persisted.

The diagnosis was delayed in our case because of the low clinical suspicion, even though her platelet counts were persistently elevated for about three years before presenting at our facility. In a study of 23,796 consecutive autopsies performed between 1970 and 1982, representing 84% of all in-hospital deaths in the Malmö city population, the population prevalence of PV thrombosis was 1% (254 patients) [6]. Among these 254 patients, only 3% (seven cases) of PV thrombosis cases were due to a Myeloproliferative Neoplasm (the number of patients with ET was not specified). This shows that portal vein thrombosis due to ET is quite rare. In addition, out of the seven cases that developed PV thrombosis due to an MPN, only two developed esophageal varices. Cavernous transformation of the portal venous system, described in our case, refers to the development of a network of tortuous collateral vessels bypassing the obstructive area in the PV due to chronic occlusion. This acts as a compensatory mechanism to aid in venous return of blood to the liver once its blood supply is reduced significantly. This was described by Cai et al. [7] in a 36-year-old female in the Sir Run Run Hospital, Zhejiang Province of China, who had recurrent upper quadrant pain for one year and violent hematemesis from esophageal varices. Portovenography showed non-visualization of the portal vein with the formation of multiple collaterals in the hepatic hilum.

To the best of our knowledge, there is no consensus regarding the role of anticoagulation in patients with chronic PV thrombosis due to ET who develop esophageal varices. Therefore, the use of anticoagulants is dependent on individual risk factors such as bleeding risk and risk of progression of thrombosis. The bone marrow aspirate in our case was non-diagnostic due to a lack of marrow elements. This could have made us suspect Primary Myelofibrosis. However, the absence of fibrosis on bone marrow biopsy, lack of a leukoerythroblastosis on peripheral blood smear, and the absence of worsening cytopenias would make Primary Myelofibrosis less likely [8]. Concerning the diagnostic criteria, our patient had three of the four major criteria required for diagnosis, these included platelet counts greater than 450 x 10⁹/L, criteria for BCR-ABL positive CML not met, and demonstration of JAK2 mutation [9]. The criteria for bone marrow biopsy were not met, likely due to poor marrow sampling. In addition to primary myelofibrosis being less likely as mentioned above, we were able to rule out the possibility of post-ET myelofibrosis due to the absence of an increase in spleen size more than 5cm, absence of constitutional symptoms, and the lack of worsening anemia (> 2 g/dL decrease from baseline hemoglobin concentration) [9]. Unlike the case of a 46-year-old woman who presented with hematemesis due to esophageal varices and portal vein thrombosis in the context of ET [10], our patient did not have any episodes of hematemesis. They faced a similar dilemma in terms of whether or not to administer anticoagulation. Our patient´s varices were banded, so we felt more at ease administering anticoagulation with a direct oral anticoagulant (DOAC). After about eighteen months of follow-up, our patient did not have any episodes of hematemesis.

Conclusion     

Here we describe a late diagnosis of ET in a 46-year-old female with the uncommon complication of portal vein thrombosis and esophageal varices. She had a history of persistent thrombocytosis, which had not been properly addressed. The history and physical exam provided us with clues that could point towards the diagnosis of ET, subsequently confirmed with a positive JAK2 test and negative BCR-ABL. To the best of our knowledge, this represents the first published case of ET from Cameroon, particularly with these complications as mentioned earlier. Paying attention to any patient with persistently elevated platelet counts is important, and secondary thrombocytosis has to be ruled out. To the best of our knowledge, there is no consensus on managing portal vein thrombosis due to a myeloproliferative neoplasm in the context of esophageal varices. This is due to the high risk of bleeding from varices in the context of anticoagulation. At the same time, in the absence of anticoagulation, thrombosis can progress and cause organ dysfunction if left unaddressed. Anticoagulation in such cases should be administered with caution, ideally after high-grade varices have been banded.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Patient Management: Jervis Somo Lambi and Gerard Tchinda. Data collection: Jervis Somo Lambi, Gerard Tchinda, Frank Foin Alei, Edmond Asongtia Tintinu, Henry Zong. Manuscript Drafting: Jervis Somo Lambi, Frank Foin Alei, Edmong Asongtia Tintinu. Manuscript Revision: Jervis Somo Lambi, Gerard Tchinda and Frank Foin Alei, Henry Zong contributed to the revision and supervision of the manuscript. All authors have read and approved the final version of this manuscript.

Acknowledgement     

The authors express gratitude to the patient for her consent to publish this case for educational purposes.

Table and figures     

Table 1: CBC parameters of the patient while on hydroxyurea

Figure 1: ultrasound image demonstrating an enlarged spleen, measuring 189.3 mm in length

Figure 2: Doppler ultrasound, performed in color mode, demonstrating an absence of blood flow (blue arrow) within the portal vein

Figure 3: Doppler ultrasound, performed in color mode, demonstrating an absence of blood flow (blue arrow and yellow dashed lines) within the venous territories shown: A) splenic vein; B) superior mesenteric vein (SMV)

Figure 4: color Doppler ultrasound of the portal vein (blue arrow, yellow dotted line) illustrating cavernous transformation, where new collateral vessels have formed as a compensatory mechanism for obstruction

Figure 5: color Doppler ultrasound of the renal vein displaying splenic collaterals (blue arrow); the development of these collateral vessels represents a compensatory mechanism in response to splenic vein obstruction

References