A 45-year-old man with flank pain and inability to ejaculate



Chief Symptom

A 45-year-old man with left-sided flank pain and difficulty ejaculating.

History of the Present Illness

The patient developed left-sided flank pain in April 2010. He was evaluated in the emergency department when the pain had persisted for 3 weeks. He reported worsening of the pain in the evening and difficulty with ejaculation since the pain began: “I get a knot in my side,” then “I can't ejaculate.” He also reported associated low back pain and increased pain with bending to his left. His general physical examination was unremarkable. Urine analysis demonstrated 30 mg/dl protein, 10 mg/dl ketones, 8 mg/dl urobilinogen (reference value <2.0), 8 red blood cells/high-power field (hpf), 2 white blood cells/hpf, and mucous. He was prescribed ibuprofen for abdominal pain and was advised to follow up with his primary care provider.

The pain persisted. He saw his primary provider in July with worsening epigastric and left-sided abdominal and flank pain. By this time, he had decreased appetite, early satiety, and weight loss of 22 pounds. His primary provider ordered a computed tomographic (CT) scan of his abdomen and pelvis, which demonstrated a rind of soft tissue in the retroperitoneum surrounding the aorta and inferior vena cava and wall thickening of the sigmoid colon and rectum. A biopsy of the mass was done by interventional radiology at another hospital. Since his pain was unrelieved by acetaminophen/hydrocodone, he was admitted to our hospital for further evaluation and pain control. He did not know the results of the biopsy.

Medical History

He had been relatively healthy prior to the onset of illness in April. Atheromatous calcification of the aorta with mild ectasia inferiorly measuring up to 2.3 cm in diameter was noted incidentally on a CT scan obtained to evaluate for ureteral calculi in September 2008. He had a remote history of gonorrhea and chlamydia infections, which were treated. He had been seen in the emergency department for right testicular pain following trauma at age 35 years and left testicular pain without clear etiology at ages 40 and 43 years. The first episode of left testicular pain was attributed to epididymitis. The second was attributed to suspected passage of a ureteral calculus.

Social and Family History

The patient's family history was negative for rheumatic disorders. Several family members, including his mother, had diabetes mellitus and hypertension. His sister had thyroid disease. He was unemployed, but previously worked as a baker. He smoked one-quarter of a pack of cigarettes per day for several years, and previously smoked marijuana. He consumed, on average, 1 alcoholic beverage per week and denied intravenous drug use. He was sexually active and did not consistently use barrier protection.

Review of Systems

The patient reported persistent inability to ejaculate, without decrease in arousal or orgasm and without associated pain. He also reported low back pain that preceded his current illness by years. He had constipation for 2 weeks. He denied fevers, chills, and night sweats. He reported no epistaxis, bleeding gums, easy bruising, headaches, chest pain, palpitations, lower extremity edema, dyspnea, cough, hemoptysis, sinusitis, vision changes, nausea, vomiting, diarrhea, hematochezia, melena, dysuria, hematuria, myalgias, arthritis, or rash.

Physical Examination

On admission, his temperature was 97.5°F, his heart rate was 72 beats/minute, and his blood pressure was 101/62 mm Hg. His respiratory rate was 16 breaths/minute and his oxygen saturation was 100% on room air. He was intermittently uncomfortable from paroxysms of abdominal pain. His sclerae were anicteric and not injected. The mucous membranes were moist. The oropharynx was normal. He had a supple neck. The trachea was midline. There was no thyroid enlargement or nodules. The cardiac and pulmonary examinations were unremarkable. He had no peripheral edema. His abdomen was nondistended and soft. Bowel sounds were normal. He had mild periumbilical tenderness to palpation. At the midline, there was a 3–4-cm diameter pulsatile mass. There were no bruits. He had no hepatosplenomegaly. His flanks were nontender. His scrotum was normal; there was no tenderness or mass. He had one 0.5-cm diameter nodule along the left inguinal canal. There was no abnormal preauricular, cervical, occipital, supraclavicular, axillary, inguinal, or femoral lymphadenopathy. His joint examination was normal.

Laboratory Evaluation and Imaging Results

The patient had a white blood cell count of 9,800/μl (normal range 4,000–11,000) with 88% neutrophils (normal range 43–72%) and 13% lymphocytes (normal range 17–43%). The hematocrit was 33.5% (normal range 41–53%), mean corpuscular volume was 87.6 fl (normal range 80–100), and red blood cell distribution width index was 12.6% (normal range 11.5–14.5%). Platelet count was 463,000/μl (normal range 150,000–450,000). The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) level were 68 mm/hour (normal range 0–15) and 4.0 mg/dl (normal range 0–0.9), respectively. Electrolytes were normal. Serum calcium level was 8.2 mg/dl (normal range 8.2–10.2). Blood urea nitrogen and serum creatinine were 12 mg/dl (normal range 7–20) and 1.35 mg/dl (normal range 0.66–1.25), respectively. Estimated glomerular filtration rate was >60 ml/minute/1.73 m2. Iron level was 29 μg/dl (normal range 49–181), total iron binding capacity was 194 μg/dl (normal range 240–430), calculated iron saturation was 15% (normal range 15–50%), ferritin was 215 ng/ml (normal range 20–250), and automated reticulocyte count was 0.9% (normal range 0.6–1.9%). Peripheral smear demonstrated slight to moderate normochromic, normocytic anemia without an increase in polychromasia, no evidence of hemolysis, slight thrombocytosis without atypia, a polymorphous lymphocyte population, unremarkable neutrophil morphology, and no circulating blasts or atypical cells. Aspartate aminotransferase was 76 units/liter (normal range 0–55), alanine aminotransferase was 12 units/liter (normal range 0–69), alkaline phosphatase was 107 units/liter (normal range 38–126), total bilirubin was 0.3 mg/dl (normal range 0.2–1.3), total protein was 6.9 gm/dl (normal range 6.3–8.2), and albumin was 3.2 gm/dl (normal range 3.5–5.0). Serum protein electrophoresis revealed no monoclonal paraprotein. Total IgG was 1,170 mg/dl (normal range 694–1,618), IgG1 was 611 mg/dl (normal range 382–929), IgG2 was 458 mg/dl (normal range 241–700), IgG3 was 97 mg/dl (normal range 22–178), and IgG4 was 28.6 mg/dl (normal range 4–86). Thyroid-stimulating hormone was 0.614 μIU/ml (normal range 0.465–4.68). Human immunodeficiency virus antibody was negative/nonreactive. Antineutrophil cytoplasmic antibodies (ANCAs) were detected with myeloperoxidase (MPO) specificity at 17 units/ml (normal value <6).

CT scan of the chest, abdomen, and pelvis demonstrated no lymphadenopathy. There was no evidence of mediastinal fibrosis. The thoracic aorta had a normal appearance. The abdominal aorta was ectatic, measuring up to 2.9 cm. There was a thickened rind of soft tissue around the infrarenal segment of the abdominal aorta that extended inferiorly to encircle the proximal aspect of the iliac arteries bilaterally (Figure 1). There was mild hydronephrosis on the right and mild dilatation of the proximal right ureter to the level of the soft tissue mass. Mild concentric thickening of the wall of the sigmoid colon, descending colon, and rectum was unchanged from the previous CT scan.

Figure 1.

Axial (A) and coronal (B) views from the computed tomography scan of the abdomen and pelvis demonstrate a thickened rind of soft tissue around the infrarenal segment of the abdominal aorta that extends inferiorly and encroaches upon the proximal aspect of the iliac arteries.


A 45-year-old man presented with severe paroxysmal abdominal pain. This presentation occurred after 10 weeks of progressive flank pain, anorexia, weight loss, and inability to ejaculate. His CT scan demonstrated a retroperitoneal mass thought to be consistent with retroperitoneal fibrosis (RPF). The rheumatology service was consulted for recommendations regarding management of possible idiopathic RPF (IRPF).


A central question in this patient's evaluation is the etiology of his retroperitoneal mass. We first consider the various etiologies of RPF as causes of the mass, and then a potential mimic of RPF.


RPF, first described in 1905, is a rare disease (1–2 of 200,000) that affects middle-aged men twice to 3 times as often as women (1). RPF is characterized by the presence of fibroinflammatory tissue that typically originates just below the level of the aortic bifurcation near the pelvic inlet and spreads throughout the retroperitoneum, with extension bilaterally along the aorta and iliac vessels often involving the inferior vena cava and ureters. Ureteral obstruction is present in 80–100% of reported cases (2). Rarely, RPF continues above the diaphragm or/and down the pelvis and to the scrotum (1). Signs and symptoms result from mechanical effects (e.g., compression of the tubular structures, most frequently the ureter) and systemic illness. The most common clinical manifestation is abdominal, flank, or back pain (3). Constitutional symptoms (fatigue, low-grade fever, nausea, and anorexia) can be insidious, may precede or coexist with mechanical symptoms, and vary in duration. The clinical presentation and location of our patient's retroperitoneal mass was in keeping with a diagnosis of RPF. However, one needs to distinguish IRPF from secondary RPF that requires treatment directed to the underlying cause.

Secondary RPF

RPF may be secondary to the use of certain drugs, malignancy, infections, radiotherapy, surgery, trauma, asbestos exposure, histiocytosis, or amyloidosis (3–17). Medications associated with RPF include ergot alkaloids, dopamine agonists, beta-blockers, hydralazine, and analgesics (e.g., aspirin and phenacetin) (3–8). Phenacetin is no longer used medicinally, but has been used as a cutting agent to adulterated cocaine (18). The pathogenesis of RPF in these cases is unknown, but is suspected to be mediated by serotonin (3). Retroperitoneal metastasis and retroperitoneal primary tumors prompt an exuberant desmoplastic response. Carcinoids may also induce RPF by the release of several growth factors and possibly through overproduction of serotonin. Histiocytoses, including Erdheim-Chester disease (polyostotic sclerosing histiocytosis) and inflammatory malignant fibrous histiocytoma, are rare causes of RPF, considered when features of RPF are atypical or if a patient does not respond to treatment with corticosteroids (3). Their exclusion requires careful histologic examination.

The appearance of most secondary forms of RPF is nearly identical to that of so-called idiopathic disease. However, when secondary to malignant disease, the mass lesion is usually more irregularly shaped and atypically localized (3). In malignant disease, careful histologic examination reveals neoplastic cells interspersed within the abundant fibrous tissue and often invasion and disruption of neighboring muscle or bone. A monoclonal inflammatory infiltrate is compatible with lymphoma. Contiguous spread to the vertebrae or other soft tissues and/or the presence of granulomas suggest underlying infection (tuberculosis, actinomycosis, and histoplasmosis). Hemosiderin deposits may be indicative of trauma (accidental or iatrogenic) (3).

The patient had not been taking any of the medications associated with RPF and has no history of cocaine use. He lacked systemic features suggestive of a metastatic carcinoid. CT demonstrated no evidence of a primary tumor in the kidneys, adrenals, pancreas, or intestine, and no evidence of metastatic disease. Although he had weight loss, anemia, and elevations in inflammatory markers suggesting the possibility of a chronic infection, he had no clinical evidence of a focus of infection and had no evidence for contiguous spread, such as a spinal or paraspinal abscess on CT. He has no history of radiotherapy, trauma, or major abdominal surgery, and had no evidence of retroperitoneal fluid collection or hematoma on imaging.


Most cases of RPF have been considered idiopathic. The pathogenesis of IRPF may be related to chronic periaortitis (3, 19, 20). IRPF, inflammatory abdominal aortic aneurysms, and perianeurysmal RPF have similar histopathologic characteristics, including advanced atherosclerosis, medial thinning, and adventitial and periadventitial inflammation (3). Our patient demonstrated evidence of 2.3-cm abdominal aortic dilatation without a history of hypertension just over 1.5 years prior to presenting with associated RPF. Studies suggest that chronic periaortitis results from a local inflammatory reaction to antigens in the atherosclerotic plaques of the abdominal aorta (3, 21–26). However, the presence of constitutional symptoms, high concentrations of acute-phase reactants, and frequent association with autoimmune diseases that involve other organs would seem to indicate that chronic periaortitis (including IRPF) may be a manifestation of a systemic autoimmune or inflammatory disease (3, 22, 23, 27).

Vasculitis of the adventitial aortic vasa vasorum and periaortic retroperitoneal small vessels is a frequent finding in IRPF; this could promote medial thinning and aneurysm formation (4). The disease is associated with HLA–DRB1*3, an allele linked to other autoimmune diseases (diabetes mellitus, myasthenia gravis) (28). Abnormal antibodies may suggest the presence of an associated connective tissue disease or vasculitic syndrome, but may be nonspecific. Antinuclear antibodies are detected in as many as 60% (3). Other nonspecific findings include elevated ESR and CRP level, which are present in 80–100% of patients (3), and mild to moderate anemia, which is common.

A diagnosis of IRPF offered a unifying explanation for our patient's symptoms and findings, including the confluent mass around the infrarenal aorta; associated abdominal aortic aneurysm; hydronephrosis and dilatation of the ureter; increased creatinine; thickened wall of the sigmoid colon, descending colon, and rectum; abdominal pain; bowel dysfunction; elevated ESR and CRP level; thrombocytosis; normochromic, normocytic anemia; and weight loss. Our patient demonstrated low-titer ANCA with MPO specificity, but had no other evidence of ANCA-associated vasculitis. The thickened portion of the sigmoid colon was the distal portion, which is retroperitoneal. Compression of both vas deferens, which pass through the retroperitoneal space in close proximity to the rectosigmoid colon, offered the most likely explanation for his inability to ejaculate. To our knowledge, anejaculation related to IRPF has not been described previously.

Retroperitoneal Fibromatosis

A potential mimic of RPF, retroperitoneal fibromatosis, was considered but thought to be unlikely. Retroperitoneal fibromatosis is a desmoid tumor, otherwise known as a deep fibromatosis. Thirty percent to 50% of desmoid tumors arise in the abdominal cavity. The etiology is poorly understood. When associated with familial adenomatous polyposis, the condition is called “Gardner syndrome.” Up to 30% of cases of deep fibromatosis have a history of trauma. Histologically, desmoid tumors present with bundles of spindle cells in a collagenous stroma with fibroblasts concentrated at the periphery, low cellularity, and absent or rare mitotic figures or nuclear pleomorphisms. Although considered benign, deep fibromatoses are characterized by local aggressive behavior and a tendency to relapse (29). Constitutional symptoms, such as anorexia or weight loss, are typically absent.


The rheumatology service retrieved slides from the outside hospital and reviewed the result in detail with the pathologist. The biopsy sample demonstrated fibrofatty and lymphoid tissue with evidence of mixed acute and chronic inflammation (Figures 2A and B). Plasma cells were abundant. There were scattered CD20+ B cells (predominantly in follicles/nodules) and more numerous CD3+ T cells (predominantly in an interfollicular distribution pattern). CD30 and CD15, bc1-1, and pankeratin stain were negative. Immune flow cytometry was unable to be performed. Additional tissue was sent for a diagnostic test. The result was pending at the time of discharge. The patient's periaortic retroperitoneal mass was therefore thought to be consistent with IRPF. He received high-dose corticosteroid treatment and was discharged on prednisone 1 mg/kg/day with a plan for close followup with urology and rheumatology. He did not require ureteral stenting.

Figure 2.

Patient's biopsy specimen. Low-power (A) and high-power (B and C) views of the specimen with hematoxylin and eosin (A and B) and immunohistochemical staining for IgG4 (C) demonstrate fibrofatty and lymphoid tissue with a lymphoplasmacytic infiltrate. Numerous plasma cells stain positive for IgG4.


The diagnostic test was special staining for IgG4 on the biopsy specimen. On the basis of the abundance of plasma cells (similar to biopsies seen in cases of autoimmune pancreatitis), we suspected IgG4-related systemic disease. Staining for IgG4 was positive (Figure 2C). In the areas corresponding to IgG4-positive density, IgG-positive cells were approximately 60/hpf with an IgG4/IgG ratio of 20%.


IgG4-related systemic disease is an inflammatory and fibrosing condition that may affect many organs and is characterized histologically by lymphoplasmacytic inflammation with IgG4-positive cells and exuberant fibrosis (30, 31). The disease was first described in the context of pancreatitis in 1961 (32). IgG4-related pancreatitis became more widely recognized as an autoimmune condition in 1995 (33). Since the first report by Kamisawa et al in 2003 (34), extrapancreatic involvement has become increasingly recognized.

Many different terms have been used to describe conditions that have become more recently recognized as members of a spectrum of clinical involvement in IgG4-related systemic disease (31, 35–38). Full recognition of the disease is challenging, as involvement of different organs affected is often metachronous. Neild et al report their results of a comprehensive literature review on disease associations with idiopathic systemic fibrosing conditions (pseudotumor, myofibroblastic tumor, plasma cell granuloma, systemic fibrosis, xanthofibrogranulomatosis, and multifocal fibrosclerosis) and identify reports associating those conditions with organ-specific conditions (such as sclerosing pancreatitis, IRPF, Reidel's thyroiditis, panniculitis, Weber-Christian syndrome, and retro-orbital tumor) (31). They also include associations with a range of autoimmune diseases. Knowledge of these potential synonyms of IgG4-related systemic disease and associations can help identify IgG4-related disease that may otherwise be overlooked.

In IgG4-related systemic disease, serum IgG4 is often elevated, but may be normal. Furthermore, the degree of IgG4 staining does not necessarily correlate with elevated serum IgG. ESR and CRP level may be monitored during acute phases of disease, but are nonspecific (3). CT is most often used to determine the extent of disease and to monitor disease over time. Fluorodeoxyglucose–positron emission tomography allows for assessment of the metabolic activity of the mass, determination of the full extent of vascular inflammatory involvement, and identification of other disease sites (such as those seen in multifocal fibrosclerosis) (3, 39). Combined positron emission tomography–CT imaging offers the ability to assess both the extent and metabolic activity of the mass and therefore, when available, is useful for monitoring response to treatment.

IRPF should be considered a potential manifestation of IgG4-related systemic disease. Neild et al reviewed the histology of biopsy samples from 12 patients with a diagnosis of IRPF seen at their hospital over the previous 10 years and examined the sections for evidence of IgG4-expressing plasma cells (31). There was a significant increase in IgG4-positive plasma cells compared with controls in all cases. Even when few plasma cells were seen, the majority expressed IgG4. In comparison, none of the wide range of control sections showed significantly increased numbers of IgG4-positive cells. Clinicopathologic evaluation of 17 patients with IRPF in Japan demonstrated that IRPF may be classified into IgG4-related and non–IgG4-related cases. The study identified 10 cases as IgG4 related, 5 of which had sclerosing lesions at other sites. All patients with IgG4-related IRPF were male; all except 1 of the patients with non–IgG4-related IRPF were female (40).

Stone et al found that IgG4-positive lymphoplasmacytic infiltrate is present in cases of aortitis (37). During a 5-year period at Massachusetts General Hospital, 5.2% of thoracic aortitis cases were noninfectious. Of those, IgG4-related systemic disease accounted for 9% and accounted for 75% of lymphoplasmacytic aortitis cases (38). In keeping with the theory of aortitis in the pathogenesis of IRPF, the findings of Stone et al lend credence to a theory that IgG4-related RPF may be related to a chronic lymphoplasmacytic aortitis.

The role of IgG4 in IgG4-related systemic disease is unclear. IgG4 normally comprises only 3–6% of total IgG, has a low affinity for the target antigen, is unable to bind C1q complement, and is Th2 dependent (41–43). During acute phase(s) of IgG4-related systemic disease, IgG4 is typically elevated. IgG4 is also increased in atopic dermatitis, filariasis, helminthic disease, pemphigus vulgaris, and pemphigus foliaceus (31, 41, 42, 44, 45). A feature of successful immunotherapy in allergic individuals is the induction of the allergen-specific IgG4 antibody (46). IgG4 antibody production in IgG4-related systemic disease does not appear to be monotypic: in all of the patients of Nield et al, the κ and λ chains showed a polytypic pattern (31). Studies on the pathogenesis of IgG4 disease could elucidate the role of immune-mediated mechanisms and provide new clues for treatment.

To date, there are no controlled therapeutic trials for treatment of IRPF or IgG4-related systemic disease. The available evidence for IRPF has supported a combination of initial surgical or urologic intervention along with early corticosteroid therapy, followed by steroid-sparing treatment. Patients respond well to corticosteroid therapy while the disease is in the acute phase, but not when the disease is in the chronic phase (30). Case reports of IRPF reviewed by Swartz describe sustained and effective steroid-sparing treatment after early corticosteroid therapy with cyclophosphamide, azathioprine, colchicine, mycophenolate mofetil, or tamoxifen (2). De Socio et al report success with the use of colchicine in 4 cases of RPF associated with autoinflammatory diseases (familial Mediterranean fever; recurrent pericarditis with orbital, mediastinal, and RPF; ankylosing spondylitis; and Behçet's disease) (47). Frequently, however, the response to disease-modifying antirheumatic drugs is incomplete. B cell depletion is known to be effective for pemphigus vulgaris, which is mediated by IgG4 autoantibodies (48, 49). Khosroshahi et al report a case series of successful use of rituximab for IgG4-related systemic disease (50).


The patient's pain initially resolved on corticosteroid therapy, but returned when he ran out of medication. He was restarted on prednisone and methotrexate was added by his primary rheumatologist. Magnetic resonance (MR) angiogram of the abdomen and pelvis 6 months after diagnosis showed neither worsening nor improvement in the lesions, although he was symptomatically better. Azathioprine was added, as he was unable to taper off prednisone despite methotrexate. MR angiogram 9 months after presentation showed the lesions had improved. Repeat urologic examination found no alternative cause for his anejaculation.

Repeat ANCA testing 2 months later was negative. The ANCA with MPO specificity in our patient appears to have been nonspecific. An association of ANCA with IgG4-related systemic disease has not been reported. Screens for involvement of IgG4 disease at other sites have been unrevealing.


Our patient presented with classic signs and symptoms of IRPF, including both functional and systemic symptoms, and was found by biopsy to have IgG4-related systemic disease. IRPF should be regarded as part of the spectrum of site involvement possible in IgG4-related systemic disease. Abnormal antibodies may suggest the presence of an associated connective tissue disease or vasculitic syndrome, but may be nonspecific. To our knowledge, this is the first report of anejaculation related to IgG4-related systemic disease. The distinction of IRPF as IgG4 related or not in a given patient allows for better prediction of the possibility of other IgG4-related processes at other anatomic sites, and should inform our treatment approach.


IgG4-related systemic disease.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Gertner had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Muskardin, Gertner.

Acquisition of data. Muskardin, Gertner.

Analysis and interpretation of data. Muskardin, Gertner.