Spontaneous splenic rupture due to extramedullary haematopoiesis in a patient with inherited thrombocytopenia

INTRODUCTION

Advances in the knowledge of inherited thrombocytopenias (ITs) achieved since the beginning of this century have changed our view of these disorders, as it has become clear that reduced platelet count is not the only clinical issue of affected subjects¹. In fact, patients with ITs can not only present with a number of other congenital defects, but also have the predisposition to acquire additional severe haematologic diseases, such as leukaemia or bone marrow aplasia, or extra-haematologic disorders¹. For example, patients with MYH9-related disease (MYH9-RD), the most frequent form of IT, are predisposed to develop kidney failure, deafness, juvenile cataracts, and alteration of liver enzymes^1,2.

Recent reports prompted the hypothesis that extramedullary haematopoiesis (EMH) could be another acquired manifestation of ITs^3,4. Here, we report a patient with MYH9-RD and massive splenic EMH complicated by spontaneous splenic rupture.

CASE REPORT

A 35-year old male presented with chronic severe thrombocytopenia (platelet count 20×10⁹/L) and mild bleeding tendency. He reported a lifelong history of mild easy bruising as the only bleeding symptom. He had previously undergone two dental extractions and one major surgical procedure (excision of branchial cyst), which were carried out after the administration of tranexamic acid or desmopressin without any bleeding complication. After diagnostic workup for ITs, a diagnosis of MYH9-RD due to the p.Lys74del in MYH9 was made. This variant is associated with moderate-to-severe thrombocytopenia and high risk of sensorineural deafness^2,5. In fact, from his late 30s, the patient had developed bilateral hearing loss. The clinical picture remained stable until the age of 47, when he began to present frequent life-threatening bleeding episodes requiring hospitalisation, including spontaneous relapsing subdural haemorrhages, and severe recurrent gastrointestinal bleeding from duodenal erosions (Helicobacter pylori-negative) that required massive red blood cell transfusions. At that time, we found that the patient’s platelet count had decreased from its usual value (around 20×10⁹/L) to around 5×10⁹/L; he also presented moderate anaemia and mild neutropenia. Platelet function, investigated by flow cytometry as described⁶, was normal, suggesting that the decrease in platelet count was the only reason for the worsening of the bleeding tendency. Physical examination and a computed tomography (CT) scan demonstrated massive splenomegaly (cranio-caudal length: 25 cm). Re-examination of the patient’s clinical records showed that actually a mild splenomegaly had already been described 5 years earlier. Of note, thrombocytopenia was found to be refractory to platelet transfusions that were given in order to manage the bleeding episodes, in spite of the absence of detectable platelet allo-antibodies. We then treated the patient with the thrombopoietin-mimetic eltrombopag (75 mg/day for 30 days), which is usually effective in MYH9-RD^6–8, but also this treatment failed to increase platelet count. A diagnostic work-up excluded the most common causes of splenomegaly, and a total body CT scan and a positron emission tomography (PET)-CT scan excluded additional masses at other sites. A splenic puncture or splenectomy was therefore considered. However, these interventions were excluded given the severity of thrombocytopenia and the failure to increase platelet count. The patient’s clinical picture remained relatively stable for about two years; a close follow-up documented stability of splenomegaly and of peripheral cytopenias.

At the age of 50, the patient suddenly presented with acute anaemia (haemoglobin [Hgb] 4 g/dL) and abdominal discomfort. A CT scan documented splenic rupture with intraperitoneal haematoma (Figure 1A and B). He did not report any traumas that could explain the splenic rupture, which was therefore classified as spontaneous. Upon premedication with two apheresis platelet units, recombinant activated factor VII, and tranexamic acid, an urgent splenectomy was carried out without excessive bleeding or other complications. Histological examination of the spleen revealed EMH as the only pathological finding (Figure 1C–G). After splenectomy, platelet count increased up to the patient’s historical value (from 5 to 22×10⁹/L); we also observed an increase in neutrophil count (from 1 to 4×10⁹/L) and Hb (from 9 to 11 g/dL) (Table I). To date, one year after splenectomy, the patient is well; platelet count is stable at around 20×10⁹/L with no other cytopenias; of note, he has not suffered from any further bleeding episode and reports only mild easy bruising.

Figure 1.

Imaging of spontaneous splenic rupture, and histological demonstration of splenic extramedullary haematopoiesis

The portal-phase abdominal computed tomography (CT) scan (A, para-coronal plane; B, sagittal plane) showed massively enlarged, inhomogeneous spleen with multiple, hypodense areas and a prominent laceration at the upper pole (A, red arrow). Subcapsular haematoma (B, red arrow) and signs of peritoneal effusion (A and B, yellow arrows) were also evident. The parenchyma of the spleen presented with expanded and congested red pulp and partly ectatic sinuses, which contained aggregates of cells from the three haematopoietic lineages (C and D: Hematoxylin Eosin, 5× and 20×, respectively). Immunohistochemical staining highlighted mature megakaryocytes (E: CD61, 20×), myeloid (F: myeloperoxidase, 20×) and erythroid (G: CD71, 40×) precursors at different stages of maturation. No morphological alterations and/or cytological specific changes affecting the three haematopoietic lineages were observed, and no blasts were evident by CD34 and CD117 staining (not shown).

Table I.

Main blood cell count parameters of the reported patient

	Historic, baseline value	Massive splenomegaly	Upon splenic rupture	Two months after splenectomy	Today (one year after splenectomy)
PC (×109/L)*	20	5	4	22	25
Hb (g/dL)#§	13	9	4	11	12.9
MCV (fL)#	90	87	85	89	89
Reticulocyte count (×109/L)#	50	90	Nd	60	50
WBC/neutrophils (×109/L)#	5/3	2.5/1	3.9/2.4	7/4.3	7.8/5.5

Blood cell counts at baseline (average of historic values before the development of splenomegaly), at the time of the development of massive splenomegaly, at the time of the spontaneous rupture of the spleen (severe haemorrhagic anaemia and mild increase of white blood cell [WBC] count due to inflammatory response), two months after splenectomy, and to date (i.e., one year after splenectomy).

^*As determined by microscopic platelet count.

^#As determined by an automated cell counter.

^§To investigate the characteristics of anaemia, the concentration of haptoglobin, lactic dehydrogenase, and bilirubin were also assessed; all these parameters were within the normal ranges. Moreover, the direct Coombs antibody test returned negative results on two different assessments.

PC: platelet count; Hb: haemoglobin concentration; MCV: mean corpuscular volume; Nd: not determined.

To investigate the possible mechanisms of EMH in our patient, we measured circulating CD34-positive progenitor cells. CD34-positive cells were 5.74/μL (0.09% of leukocytes), i.e., slightly increased compared to healthy subjects⁹. Further studies are needed to interpret the possible significance of this finding. We also measured the patient’s serum thrombopoietin level, which was found to be within the normal range on two different occasions (45.7 and 70.1 pg/mL, respectively)⁶.

DISCUSSION

The first reported patient with IT and EMH was a 55-year old female with MYH9-RD who had massive paravertebral EMH³. The second report concerned a 67-year old woman affected with ANKRD26-related thrombocytopenia and presenting a large pelvic mass consisting of EMH^4,10.

In both these cases, the large amount of ectopic haematopoietic tissue was asymptomatic; its identification was incidental and raised the suspicion of a neoplastic disease¹¹. Conversely, in the case reported here, splenic EMH has been asymptomatic for a long time but eventually led to acute spontaneous rupture of the spleen, which seriously endangered the patient’s life. Moreover, in view of the changes in blood counts before and after splenectomy, it is likely that the massive splenomegaly caused the reduction of platelet count from the patient’s baseline value to 5×10⁹/L, leading to the occurrence of severe, recurrent spontaneous bleeding episodes. In fact, the patient had never presented significant bleeding events before the age of 47. Finally, we also hypothesise that a mechanism of platelet sequestration in the massively enlarged spleen was responsible for the refractoriness to platelet transfusions and the failure of platelet count to increase after eltrombopag treatment, two interventions that are usually highly effective in MYH9-RD patients^1,6–8.

A feature common to the three reported patients with EMH was the severity of thrombocytopenia since birth. From this point of view, it is interesting to refer back to the 1974 report of EMH being present in 48 out of 50 spleens of patients who were splenectomised for immune thrombocytopenia (ITP)¹². A more recent investigation of the largest EMH case series reported so far concluded that, excluding myeloproliferative neoplasms, ITP was one of the most frequently associated conditions¹³. It is therefore conceivable that the lifelong, severe thrombocytopenia of the patients discussed here had a causative role in their massive EMH. This hypothesis is also supported by the fact that the three patients were all over 45 years old. In conclusion, our observation provides further support to the hypothesis that some IT patients with severe thrombocytopenia may develop massive EMH at a mature age. Although we are unable to provide any mechanistic explanation, this phenomenon could be clinically relevant. Moreover, EMH should be considered in the differential diagnosis of splenomegaly and thoracoabdominal masses of unclear nature in patients with IT, particularly in the absence of general systemic symptoms or signs of infiltration.