Final Diagnosis -- A 79-year-old male with postoperative spindle cell nodule (PSCN)




This is a spindle cell lesion located in the pancreas in close proximity to the duodenal wall. Given the microscopic appearance and the location, the differential diagnosis includes leiomyosarcoma, gastrointestinal stromal tumor (GIST), inflammatory pseudotumor (IPT), and postoperative spindle cell nodule (PSCN).

Leiomyosarcoma has to be considered in the evaluation of spindle cell lesions. The retroperitoneal location of the current lesion raised that possibility, though the lesion appeared to be largely confined to the substance of the pancreas. Leiomyosarcoma consists of malignant smooth muscle cells; the diagnostic criteria include atypia, mitosis and/or coagulative tumor cell necrosis. In addition the age of the patient, the size of the tumor, the pattern of tumor margins, and the presence of vascular invasion should be taken into account before making the diagnosis [1] Leiomyosarcoma is positive for desmin, alpha smooth muscle actin, p53, and MIB-1. [2]

Lack of atypia, mitosis, and necrosis effectively excluded leiomyosarcoma from the differential diagnosis in our case.

Gastrointestinal Stroma Tumor was considered in the differential because of the proximity of the lesion to the duodenal wall. It is the most common nonepithelial neoplasm of the stomach and small intestine; it can be benign, intermediate or malignant. The origin of this tumor remains a mystery; some pathologists still believe that the interstitial cell of Cajal is the cell of origin. The tumor is positive for c-kit, CD34, usually exhibits positivity for smooth muscle markers, and sometimes S-100. In this case, the negativity of CD34 and c-kit ruled out GIST from the differential diagnosis. [3]

Inflammatory Pseudotumor (IPT): The term inflammatory pseudotumor first appeared to describe lesions similar to neoplastic conditions at the clinical, macroscopic and microscopic level, all these lesions were thought to be of reactive nonneoplastic nature. Later, this term was restricted to describe spindle cell proliferation against a heavy inflammatory infiltrate composed predominantly of lymphocytes and plasma cells. [4] Some authors divided he development of IPT into early and late stages. In the beginning, an obvious mixture of acute and chronic inflammation occurs on a fibroblastic background. Later dense distribution of fibroblasts predominates in a background of collagen and chronic inflammation. [5] Recently, it has been postulated that IPT is a neoplastic condition and clonal chromosomal abnormalities involving 2p22-24, [6] and anaplastic lymphoma kinase (ALK) rearrangement and expression restricted to the myofibroblastic component have been demonstrated. [7] Some of these lesions behave more aggressive and have been classified as a low-grade sarcoma, [6] with the capability to recur, [8] infiltrate, [9] and metastasize. [10]

Thus, the entity of inflammatory myofibroblastic tumor (IMT) emerged. The subject became more complicated when recent studies separate inflammatory pseudotumor from inflammatory myofibroblastic tumor of the Pancreas, assuming that the previous is a reactive phenomenon and the latter is a true neoplasm. They also attributed the aggressiveness of some IPTs to the development of an intense and self-maintaining immune response as a result of autoimmune disturbance; this immune response was the resource of connective tissue growth factor that lead to recurrence. [11] World Health Organization (WHO) continues to classify IMT as a distinct borderline lesion with uncertainty as to whether it is reactive or neoplastic in nature. [7, 12, 13] IMT is usually idiopathic. However, previous trauma and immunosuppression may play a causative role. [7] The lung, liver, gastrointestinal tract are the most common sites of origin, [5] while pancreatic lesions are very rare. [14] Immunohistochemically: the lesion is diffusely positive for vimentin, has variable positivity for smooth muscle actin and desmin. Cytokeratin staining is usually focally positive to absent. [7]

Morphologically, this particular case shares a lot of features with IMT taken into account the edematous background, low to medium cellularity, and the absence of mitotic figures. However, the lack of pleomorphism, inflammation or mitoses, occurrence after surgery, and associated findings of traumatic neuroma and suture granuloma argue against the diagnosis IMT.

Postoperative Spindle Cell Nodule (PSCN):
This is a reactive, non-neoplastic lesion consisting of closely packed, usually mitotically active spindle cells. The superficial portion of the lesion may look like granulation tissue. However, the deeper portion may have a sarcoma-like appearance simulating leiomyosarcoma but without significant pleomorphism. Often, it contains numerous extravasated red blood cells resulting in Kaposi's sarcoma-like appearance. This lesion develops several weeks to several months following surgical intervention at the site of the operation. [12, 3]

Postoperative spindle cell nodule develops usually in the bladder, prostatic urethra, or vagina. Its mitotic rate varies between 1 to 25 per 10 hpf. No atypical mitoses are identified. [3]

Immunohistochemically: the lesion is vimentin, desmin, and actin positive, and EMA negative. It shows strong and unexplained low molecular weight keratin positivity. [3] This lesion differs from IMT in the following: it is smaller in size, has less myxoid stroma, higher degree of cellularity, less pleomorphism, and higher tendency for keratin positivity. [3]

Electron microscopy demonstrates that the spindle cells in the inflammatory myofibroblastic tumors have more developed myofibroblastic features - prominent rough endoplasmic reticulum, modesty developed myofilaments- than postoperative spindle cell nodule. [15]

Our case had features consistent with the diagnosis of PSCN such as spindle cell proliferation, extravasated red blood cells, lack of pleomorphism, reactive nature, and occurrence after surgery.

We consider this lesion to be reactive in nature, given that it occurred after a surgical procedure and is associated with other reactive phenomena (traumatic neuroma and giant cell reaction to suture material).

The absence of mitosis and the low cellularity may be explained by the chronicity of the lesion; it was discovered 3 years post-surgery, while the other described postoperative spindle cell nodules usually presented weeks to months after surgery. [12, 3]

To our knowledge, no other cases of pancreatic post-operative spindle nodule have been described in the literature.


  1. Stanley J. Robboy, Malcolm C. Anderson, Peter Russell. In "Pathology of the Female Reproductive Tract". pp 402-410. Churchill Livingstone, London UK. 2002.
  2. David J. Dabbs. In "Diagnostic Immunohistochemistry" pp 86-88 and 498-499. Churchill Livingstone, Philadelphia 2002.
  3. Rosai, J. in "Rosai and Ackerman's Surgical Pathology" 9 ed. pp 674-678, 1324, 1325, 1367, 1515. Mosby, New York. 2004.
  4. Nonaka D, Birbe R, Rosai J. So-called inflammatory myofibroblastic tumour: a proliferative lesion of fibroblastic reticulum cells? Histopathology. 2005 Jun;46(6):604-13. Review.
  5. Barnes L. in "Surgical Pathology of the Head and Neck" 2 ed. pp 2148, 2149. Marcel Dekker, Inc. New York. 2001.
  6. Fisher C. Low-grade sarcomas with CD34-positive fibroblasts and low-grade myofibroblastic sarcomas. Ultrastruct Pathol. 2004 Sep-Dec;28(5-6):291-305. Review.
  7. Barnes L, Eveson J, Reichart P, Sidransky D. WHO Classification of Tumors. Pathology & Genetics. Head and Neck Tumors. pp 150-151. IARCPress. Lyon, France. 2005.
  8. Coffin CM, Watterson J, Priest JR, Dehner LP. Extrapulmonary inflammatory myofibroblastic tumor (inflammatory pseudotumor). A clinicopathologic and immunohistochemical study of 84 cases. Am J Surg Pathol. 1995 Aug;19(8):859-72.
  9. Browne M, Abramson LP, Chou PM, Acton R, Holinger LD, Reynolds M. Inflammatory myofibroblastic tumor (inflammatory pseudotumor) of the neck infiltrating the trachea. J Pediatr Surg. 2004 Oct;39(10):e1-4.
  10. Morotti RA, Legman MD, Kerkar N, Pawel BR, Sanger WG, Coffin CM. Pediatric inflammatory myofibroblastic tumor with late metastasis to the lung: case report and review of the literature. Pediatr Dev Pathol. 2005 Mar-Apr;8(2):224-9. Epub 2005 Mar 8.
  11. Esposito I, Bergman F, Penzel R, di Mola FF, Shrikhande S, Buchler MW, Friess H, Otto HF. Oligoclonal T-cell populations in an inflammatory pseudotumor of the pancreas possibly related to autoimmune pancreatitis: an immunohistochemical and molecular analysis. Virchows Arch. 2004 Feb;444(2):119-26. Epub 2004 Jan 14.
  12. Tavassoli F, Devilee P. WHO Classification of Tumors. Pathology & Genetics. Tumors of the Breast and Female Genital Organs. pp 92, 93, 283, 305. IARCPress. Lyon, France. 2003.
  13. Travis W, Brambilla E, Muller-Hermelink K, Harris C. WHO Classification of Tumours. Pathology & Genetics. Tumor of the Lung, Pleura, Thymus and Heart. IARCPress. Lyon, France. 2004.
  14. Nakamura Y, Inui K, Yoshino J, Tokoro T, Sabater L, Takeda S, Yamashita K, Okochi O, Nakao A. Inflammatory myofibroblastic tumor (inflammatory fibrosarcoma) of the pancreas: a case report. Hepatogastroenterology. 2005 Mar-Apr;52(62):625-8.
  15. Eyden B. Electron microscopy in the study of myofibroblastic lesions. Semin Diagn Pathol. 2003 Feb;20(1):13-24. Review.

Contributed by Ibrahim Batal, MD and Teresa McHale, MD

Case IndexCME Case StudiesFeedbackHome