Cases
A 35-year-old female patient presented with chronic mild pain in the left flank without fever, nausea, vomiting or weight loss. There was no prior history of alcohol or tobacco abuse and surgery. She was in continuous use of progesterone pills due to a prior diagnosis of endometriosis. Physical examination revealed a symmetric, soft and non-tender abdomen. Complete blood count and biochemistry laboratory tests were within normal range.
The patient had an abdominal MRI performed during the endometriosis investigation, showing a 2-cm nodule in the tail of the pancreas. It had a hyperintense signal on the T2-weighted sequence and an enhancement in the arterial phase, an appearance suggestive of a PNET. We proceeded with the workup with an abdominal CT and technetium-99m scintigraphy to rule out an AS [ Figures 1 and 2 ]. Neither one was compatible with AS, thus increasing the hypothesis of PNT.
The above abdominal scan in an axial sequence (a-d) shows a hypervascular nodule in the pancreatic tail
99-m Technetium scintigraphy does not show accessory spleen
The findings were discussed in the multidisciplinary tumour board and were planned for laparoscopic distal pancreatectomy with splenectomy. The patient was adequately vaccinated. The procedure was performed with a minimally invasive approach, and the patient evolved with Type-B pancreatic fistulae and managed conservatively. Pathology [ Figures 3 - 5 ] was compatible with a projection of ectopic splenic tissue into the pancreas, separated by a fibrous capsule from the pancreatic parenchyma, without atypia. She is asymptomatic and with no evidence of disease in a 2-year follow-up.
Macroscopic view of the tail of the pancreas with attached spleen
Macroscopic/gross pathology of the intrapancreatic lesion with attached spleen
Intrapancreatic accessory spleen tissue separated from the pancreatic tissue by a capsule structure (H and E, ×10)
A 72-year-old female patient presented due to an incidental finding of a nodule in the tail of the pancreas during the workup for a laparoscopic cholecystectomy. Her only complaint was pain in the right upper quadrant after a meal. Physical examination and laboratory work were normal. Investigation proceeded with a CT scan that showed a 3-cm hyperdense nodule in the pancreatic tail [ Figure 6 ]. To confirm the suspected diagnosis of an AS and rule out the possibility of a PNET, an EUS-guided biopsy was performed and established the diagnosis of ectopic spleen tissue.
Another example of the tomographic aspect of an accessory spleen in the pancreatic tail (a-c), with an enhancement very similar to the spleen in the different contrast phases
Intro
The spleen is the largest lymphatic organ in the body, derived from embryologic mesenchymal cells. During the migration of these cells along the splenic vessels, an accessory spleen (AS) may appear if the splenic buds are not completely fused.[ 1 ] It could also occur as an acquired condition referred to as splenosis, usually due to cell spillage after splenectomy for splenic trauma. The AS occurs in 10% of the normal population, and the most common sites include the splenic hilum, followed by the greater omentum and pancreas.[ 2 ]
Despite being uncommon, its greatest importance relies on the fact that the diagnosis can impose a particular difficulty, leading to misdiagnosis of potential malignant conditions with different therapies. The list of differential diagnoses ranges from enlarged lymph nodes to pancreatic tumours and metastasis. The correct identification of this condition also has an important role in the management of some haematological disorders that require splenectomy, when the AS must be correctly identified and removed.[ 3 ]
When the AS occurs in an uncommon location, such as inside the pancreas, the diagnosis is more complex. The intrapancreatic AS (IPAS) is usually found incidentally on cross-sectional images as a well-defined hypervascular nodule, mainly in the pancreatic tail, showing the same signal and intensity as the spleen.[ 4 ] In addition, it can potentially grow and mimic pancreatic tumours.[ 5 ] Therefore, IPAS should be included as a differential diagnosis of hypervascular solid lesions in the tail of the pancreas, along with pancreatic metastasis (especially from renal cell carcinoma), solid pseudopapillary tumour and pancreatic neuroendocrine tumours (PNETs).[ 6 ]
When an intrapancreatic lesion is discovered, especially if it is located in the pancreatic tail, advanced diagnostic methods are needed to potentially avoid unnecessary surgery for a benign condition. Magnetic resonance imaging (MRI), computerised tomography (CT), technetium-99m scintigraphy and endoscopic ultrasound (EUS)-guided fine-needle aspiration are the most valuable tools to increase the odds of an accurate diagnosis.[ 7 8 ]
We report two cases of IPAS, their diagnostic workup and treatment, emphasising the difficulty in achieving an accurate diagnosis. The approach to a hypervascular lesion in the pancreatic tail is discussed with emphasis on establishing the differential diagnosis with IPAS.
Conclusion
This review, aligned with our experience, highlights the importance of a proper investigation of solid hypervascular lesions in the pancreatic tail. We should consider IPAS as a differential diagnosis and underscore the crucial role of MRI, Tc99m, (SPIO)-enhanced MRI and occasionally EUS in confirming the diagnosis and potentially avoiding unnecessary surgeries.
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
There are no conflicts of interest.
Discussion
The pancreatic tail is the second most frequent region for AS. Often, it is an asymptomatic disease found as a pancreatic nodule during investigation for other abdominal symptoms. It is known to be an uncommon diagnosis, even though a recent increase in its incidence has been reported due to the increasing use of advanced abdominal imaging, such as CT or MRI, for non-related abdominal symptoms.[ 9 ] In abdominal imaging, AS usually presents as well-circumscribed rounded lesions with smooth contours and arterial hyperenhancement in the pancreatic tail. Thus, it can easily mimic PNET, adenoma and carcinoma.[ 10 ] Correct diagnosis and investigation may decrease the risk of unnecessary surgeries.
CT and MRI are crucial in the investigation of any solid pancreatic nodule. In the CT, the lesion presents with an identical enhancement pattern as the topic spleen. The intrapancreatic ectopic spleen usually is seen in MRI as a hyperintense lesion in diffusion-weighted/T2 and a hypointense lesion in T1-weighted phases, compared to normal pancreatic tissue. Kang et al .[ 11 ] pointed out that the isointensity of the pancreatic lesions compared with the spleen on diffusion-weighted image (DWI) was more frequently observed in IPAS than in NET, and that the mean ADC value was significantly lower in IPAS than in NET, supporting the utility of combined contrast-enhanced MRI and DWI to help differentiate PNET and IPAS.
Selective spleen scintigraphy, which is performed using heat-damaged red blood cells (HDRBC) labelled with Technetium-99m (Tc99 m), is really useful in the differential diagnosis, as it provides selective and robust splenic imaging, allowing the non-invasive diagnosis of AS. Scintigraphy can also be performed with Tc99m-colloid, which is diffusely captured by the reticuloendothelial system and therefore is less specific. However, the low resolution of scintigraphy might restrict its utility to the diagnosis of lesions smaller than 1 cm and to the ones with minimal functional splenic tissue.[ 12 ] It is worthwhile to note that IPAS can give false-positive results in somatostatin receptors scintigraphy (octreoscan) due to the presence of somatostatin receptors within the splenic tissue, and thus, this cannot be used as a definitive method in differentiating IPAS from PNET.[ 13 ]
Superparamagnetic iron oxide (SPIO)-enhanced MRI can also be used for the diagnosis of IPAS, as the SPIO is phagocytosed by the reticuloendothelial system, generating a signal drop in the IPAS similar to the one observed in the topic spleen, and it has a better space resolution than scintigraphy.[ 14 ]
EUS with biopsy provides a definitive pathological diagnosis but has limitations, such as it is examiner dependent and cannot be applied for certain locations of lesions in the pancreas. Besides, it is an invasive method, and the biopsy poses the risk of complications.[ 12 ] Interestingly, Bhutani et al .[ 15 ] described the ‘bridge sign’ to help in the non-invasive diagnosis of an IPAS, which corresponds to the visualisation of a bridge of splenic tissue between the topic and the AS in the EUS.
Although multiple imaging modalities can theoretically aid in the pre-operative differential diagnosis, no single examination is capable of confirming the diagnosis of AS and completely ruling out the possibility of a PNET, as was demonstrated by Li et al .[ 16 ] in a review that included 105 patients. In this review, 15 patients were submitted preoperatively to a Tc99m-HDRBC scintigraphy, and 13 of them had a true positive result, meaning that 2 of them (13%) had a false negative. Notably, the rate of false negative was significantly higher when Tc99m-colloid scintigraphy was used (2 false negatives out of seven patients – 28.5%). Signal drop after SPIO on T2-weighted imaging was seen in 9 out of 10 patients. More than half (55%) of the patients received surgical treatment, mainly because of a pre-operative misdiagnosis of PNET (87%). The authors conclude that a combination of different imaging modalities is encouraged to help in the clinical decision-making process.
Text is read by the "Ask this paper" AI Q&A widget below.
Extraction quality varies by source — PMC NXML preserves structure
cleanly, OA-HTML may include some navigation residue, and OA-PDF can
have broken hyphenation. The publisher copy
(via DOI)
is the canonical version.