Atraumatic Splenic Rupture due to Chronic Myelomonocytic Leukemia Treated with Partial Splenic Artery Embolization

Atraumatic splenic rupture (ASR) is a rare but life-threatening cause of acute abdominal pain in patients without traumatic history and comprises 0.1–0.5% of cases of splenic rupture [1]. Splenic rupture is frequently a clinical diagnosis; although computed tomography (CT) imaging or ultrasonography (such as identification of free fluid on a FAST exam) can be helpful in confirming the diagnosis, maintaining ASR on the differential for acute abdominal pain is essential. Timely diagnosis can be especially difficult as the classic triad of left hemidiaphragm elevation, lower lobe atelectasis, and pleural effusion is frequently absent [2].

Most patients with ASR have baseline splenic abnormalities, such as splenomegaly [3]. Common causes are malignant hematological disorders (16.4%), viral infections (14.8%), and local inflammation/neoplasm (10.9%) [3]. Timely diagnosis of splenic rupture is essential as early intervention with angiography is associated with splenic salvage and improved outcomes, and splenectomy can be associated with a mortality rate as high as 2% [4, 5].

Patients with hematological conditions are at risk for splenic abnormalities, making the diagnosis of ASR an important consideration when evaluating acute abdominal pain. The mechanism by which splenic rupture occurs in these patients has been hypothesized to be due to several etiologies [6, 7]. The most frequent pathology is infiltration of spleen by the malignant cells, causing rupture of the relatively nondistensible splenic capsule through mechanical pressure. ASR can also be caused by splenic hemorrhage related to thrombocytopenia. In addition, splenic infarcts may cause architectural changes and altered vascular resistance that predispose to rupture. It is quite likely that all three of these pathways are involved in splenic rupture in these patients.

Guidelines for management of ASR are primarily based on data from traumatic splenic rupture cases and frequently involve splenectomy in hemodynamically unstable patients, although nonoperative management (NOM) is now considered the gold standard in hemodynamically stable patients with minor to moderate lesions [8]. Here we present what we believe to be the first reported case of ASR due to chronic myelomonocytic leukemia (CMML) treated with partial splenic artery embolization (SAE).

DO was a 74-year-old man with history of papillary thyroid cancer in remission and postsurgical hypothyroidism. He was previously diagnosed with myelodysplastic syndrome, with mild peripheral cytopenias and bone marrow biopsy with normal karyotype and <5% blasts. He was initially treated with decitabine and subsequently observed by primary care. He was re-referred to hematology at our institution 13 years later for rising white blood cell count (∼30,000/μL), monocyte predominant with 1–2% circulating blasts, and thrombocytopenia (40–70,000/μL). Repeat bone marrow biopsy showed persistent disease with stable blasts, and a normal karyotype with SRSF2, ASXL1, TET2, and KRAS mutations identified on next-generation sequencing. Given his persistent monocytosis and mutations, he was diagnosed with CMML in proliferative stage, with plan to start hydroxyurea.

However, prior to initiating hydroxyurea, he presented to the emergency department with acute left upper abdominal pain with chills, malaise, and lightheadedness, without history of trauma. His vitals were significant for tachycardia with a heart rate of 99 and hypotension to 103/63, with an exam with dry mucous membranes, abdominal tenderness worst in his left upper quadrant with rebound and guarding, and splenomegaly. His initial WBC was 79.9, increased from 47.2 4 days prior, hemoglobin 13.2 g/dL, platelets 52,000/μL. A CT scan demonstrated marked splenomegaly (25 cm in greatest dimension) with a 5 × 6 cm mass of mixed densities suggestive of a hemorrhagic lesion with associated splenic capsular disruption and moderate volume free fluid in the pelvic, perisplenic, and perihepatic regions concerning for hemoperitoneum (shown in Fig. 1).

He was evaluated by surgery and interventional radiology, and ultimately taken for partial SAE. During the procedure, active extravasation was noted from the interior spleen which was successfully stopped with selective embolization. The following day, his hemoglobin had downtrended to 10.9 g/dL, after transfusion of 2 units of blood. His WBC rose to a peak of 118,000/μL (1–2% blasts), for which hydroxyurea was initiated, and his platelets increased to 100,000/μL (shown in Fig. 2).

Hospital course was complicated by acute kidney injury, hyperkalemia, hyperphosphatemia, elevated uric acid, and elevated lactate dehydrogenase, concerning for “tumor” lysis. He received one dose of rasburicase and started allopurinol (shown in Fig. 2). All parameters improved and he was discharged on day 8.

Imaging post-discharge demonstrated a decreased splenic size, with a large cystic component measuring 22 × 13 × 21 cm, likely related to post-embolization liquefaction (shown in Fig. 3). At his 2-month follow-up, the patient was doing well on hydroxyurea, without further intervention.

Overall ASR related mortality rate is 12.2%, with lower rates of mortality in patients with NOM [3]. In one review, 85.3% underwent surgery with 7.4% mortality, while 14.7% had nonsurgical conservative treatment with 4.4% mortality [3]. Seven in the conservative group underwent SAE with 0% mortality. Compared to traumatic splenic rupture, ASR patients have lower rates of organ preserving surgery or nonsurgical management, likely related to older patient age and baseline splenic abnormalities, as in our patient. In our patient, diagnosis was confirmed preoperatively, and he was appropriately evaluated for splenectomy versus SAE, a minimally invasive procedure that allows for splenic salvage. Early recognition of ASR as a cause of acute abdominal pain in patients with CMML is therefore essential.

To our knowledge, this is the only case of ASR in the setting of CMML that has been treated with SAE. In general, SAE is well tolerated, with one study finding 73.1% of patients had a positive outcome with shorter hospital stays, fewer intensive care unit days, and fewer transfusions [9]. Although 8% of patients developed delayed complications (hemorrhage, splenic abscesses, and splenic pseudocysts), most patients did well [10]. In addition, there is evidence that immunological function is preserved after SAE [11]. In the most recent World Society of Emergency Surgery (WSES) guidelines, most hemodynamically stable patients without severe lesions can by monitored or undergo angiography rather than laparotomy, assuming early CT is obtained to triage severity of lesions [8]. A laparoscopic approach is an alternative to laparotomy and SAE. While laparoscopic splenectomy is preferred for benign or planned splenectomies, there are relatively few reported cases of patients with splenic rupture who undergo this technique [12, 13]. Of those patients, most were hemodynamically stable and had failed NOM, with generally positive outcomes. While it appears that this is a safe alternative to a laparotomy, patient selection is key and not every patient is eligible. At this time, there are no consensus guidelines for ASR.

On review of the literature, only 7 prior cases of splenic rupture in CMML have been reported, all treated with splenectomy (Table 1) [14]. Of these 7 patients, outcomes were usually positive or not reported; however, splenectomy in general is often complicated by bleeding and infectious complications, especially in patients with abnormal spleens. Prophylactic splenectomy has also been considered in CMML; in one study, 3/12 patients who underwent splenectomy died as a direct result of the surgery while an additional 4 patients suffered significant post-op complications [15]. There is clearly significant mortality and morbidity associated with splenectomy in CMML patients and SAE may provide an alternative.

Post-procedure leukocytosis is common and raises concerns for CMML progression. Three patients described by Steensma et al. [15] who had prophylactic splenectomies for CMML had postoperative leukocytosis >100,000/μL requiring hydroxyurea. Tracking blast percentage therefore can be reassuring against progression or transformation of hematologic disorder. In our patient, although his leukocytosis peaked at 118,000/μL, his blast percent remained between 1 and 2% with appropriate response to hydroxyurea.

Complications in our patient included elevated uric acid, potassium, and phosphorus, raising concern for a “tumor” lysis syndrome (TLS) and were treated as such with rasburicase and allopurinol. Although the patient did have an AKI, it was thought to be in the setting of hypovolemia/hypotension, rather than directly related to TLS. It could therefore be important to monitor TLS labs (potassium, phosphorus, calcium, uric acid, and creatinine) more frequently in these patients.

Overall, this case demonstrates several key points. First, a nonsurgical splenic sparing approach like SAE is feasible for patients with CMML who experience ASR. ASR is important to consider the differential for patients with hematological malignancies presenting with abdominal pain, as early diagnosis is essential in deciding splenectomy versus splenic sparing interventions. Profound leukocytosis can be common post-procedurally in patients with CMML and can be differentiated from AML transformation through monitoring of blast percentage. Finally, monitoring of TLS laboratories post-procedurally may be recommended in such patients.

Ethics approval was not required in accordance with the local or national guidelines. Written informed consent was obtained from the patient for publication of the details of their medical case and any accompanying images.

The authors have no conflicts of interest to declare.

No funding was acquired.

Yunan Nie wrote and edited the manuscript and assembled tables and figures. Andrew Kent assisted with writing and editing the manuscript and assembled tables and figures. Minh Do assisted with writing and editing of the manuscript. Maria Amaya, Catherine Klein, and Christiane Thienelt provided feedback and edits on the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.