Abstract
A 52-year-old woman presented with headaches, difficulty with word finding and left eye blindness. MRI showed enhancing frontal dural-based masses suggestive of meningiomatosis. Biopsy and debulking revealed necrotising granulomas, without discernible micro-organisms or neoplasia; a thorough clinical work-up was negative for infection and vasculitis. A CT scan showed mild bilateral hilar lymphadenopathy, a tiny pulmonary nodule and cirrhotic liver. Her subsequent alteration of mental status was attributed to hepatic encephalopathy based upon elevated aminotransferase and ammonia levels, biopsy evidence of hepatoportal sclerosis with rare granulomas and response to lactulose. A diagnosis of neurosarcoidosis of the necrotising sarcoid granulomatosis variant (NS-NSG) with atypical systemic involvement was made. This is the fifth case report of NS-NSG clinically mimicking a neoplasm and histologically mimicking a mycobacterial infection. NS-NSG can have an atypical clinical picture including intracranial masses; a thorough work-up to exclude infectious and other non-infectious aetiologies is a prerequisite to its diagnosis.
BACKGROUND
Necrotising granulomatous lesions of the central nervous system (CNS) are aetiologically diverse. These include chronic infections such as fungal or mycobacterial, as well as non-infectious causes such as Wegener granulomatosis (WG), idiopathic pachymeningitis and neurosarcoidosis of the necrotising sarcoid granulomatosis variant (NS-NSG). An accurate diagnosis is critical to deciding the optimal treatment strategy, which can be diametrically opposite (ie, intensive antimicrobial regimen vs immunosuppressants). Therefore a systematic thorough diagnostic work-up, clinicopathological correlation and appropriate follow-up are essential elements of patient management.
In this report, we present an unusual case of granulomatous dural-based masses radiologically mimicking meningiomatosis in a middle-aged woman. A thorough investigation was negative for an infectious aetiology, and the patient was diagnosed as having NS-NSG. NS-NSG is an exceedingly rare form of sarcoidosis that has only been described a handful of times in the literature, and even more rarely when mimicking a mass lesion. Awareness of this entity, potential clinical mimics, and a correct diagnostic approach are imperative as its treatment is antithetical to antimicrobial therapy yet can be very effective.
CASE PRESENTATION
A 52-year-old African–American (AA) woman presented to an outside facility with complaints of headaches, difficulty with word finding and one episode of left eye blindness. MRI of the brain showed diffuse lobular enhancement of the dura bilaterally, which was suggestive of meningiomatosis cerebri. The patient was then referred to our institution for surgical management. The patient’s past medical history was significant for only hypothyroidism. However, her thyroid-stimulating hormone (TSH) level was tested at three separate occasions during her care and was within normal limits each time (0.98, 0.53, 2.55 mcIU/ml, reference range 0.4–4.7 mclU/ml). She was not treated for hypothyroidism during her course and a more extensive investigation was not performed. Her past surgical history included the removal of a benign uterine mass. Family history was positive for an inoperable brain tumour in her grandmother. Social history was negative for tobacco abuse, alcohol, or drugs; she was not on any regular medications.
Physical exam prior to her surgery revealed anosmia and disinhibition. Routine laboratory tests showed mild leukopoenia (4400 cells/mm3, reference range 4500–11 000 cells/mm3), anaemia (12.1 g/dl, reference range 12.5–16 g/dl) and a normal platelet count (206 000 cells/mm3, reference range 150 000–400 000 cells/mm3). A white cell differential showed absolute lymphopoenia (500 cells/mm3, reference range 1000–2800 cells/mm3). All analytes of a basic metabolic panel were within the reference range and coagulation studies including prothrombin time and partial thromboplastin time were normal.
A preoperative chest x ray showed only mild multilevel degenerative changes of the thoracic spine.
A neurosurgical exploration of the mass was performed via frontal craniotomy. An intraoperative frozen section showed necrotising granulomatous inflammation without any neoplastic cells. Given this unexpected finding and the possibility of an infectious aetiology, tissue was submitted to microbiology, and her neurosurgical procedure was modified to a limited debulking rather than resection, which she underwent without any complications. A postoperative MRI of the brain showed substantial residual disease (fig 1). However, the patient recovered postoperatively and was discharged with a follow-up appointment.
Figure 1.
Precontrast (D) and postcontrast (A) sagittal MRI images (1.5 Tesla) showing a mildly hyperintense and nodular enhancing dural-based mass lesion, respectively, both with associated mass effect. Postcontrast coronal (B) and axial (C) T1-weighted MRI images (1.5 Tesla) showing bifrontal nodular dural-based enhancement with associated mass effect and oedema.
INVESTIGATIONS
Permanent sections of the tissue submitted to pathology showed diffuse necrotising granulomatous inflammation involving fibroconnective tissue, likely representing dura, abutting adjoining focally involved and gliotic brain tissue (figs 2 and 3). Compression of small vessels by these granulomas was identified, but large vessels were unaffected. No neoplastic cells, polarisable material, significant eosinophils, foamy histiocytes, emperiopolesis, viral inclusions, or vasculitis were identified. Special stains for acid-fast bacilli (AFB), fungi and spirochetes were negative. In addition, an immunohistochemical stain for syphilis was negative.
Figure 2.
Granulomas with surrounding giant cells, including Langhans and foreign body types (H&E stain; original magnification ×10).
Figure 3.
Most granulomas demonstrated central necrosis (H&E stain; original magnification ×20).
Gram stain from a smear of the lesional brain tissue submitted to microbiology showed no organisms with rare neutrophils, and aerobic as well as anaerobic cultures were negative. An acid-fast bacilli (AFB) smear and culture were negative. Finally, fungal cultures failed to yield any growth.
Furthermore, serum rapid plasma reagin test was non-reactive, a serum Lyme disease antibody screen by enzyme immunoassay was negative, and a serum histoplasma antibody screen (performed at MiraVista Diagnostics, Indianapolis, Indiana, USA) by enzyme-linked immunosorbent assay was also negative.
Over the following 2 months, the patient was admitted several times with waxing and waning mental status. Work-up was extensive and included routine laboratory tests, which showed elevated liver enzymes (aspartate aminotransferase 103 U/litre, reference range 13–45 U/litre and alanine aminotransferase 118 U/litre, reference range 5–57 U/litre). Given these results, concern was raised regarding hepatic encephalopathy. An ammonia level was obtained that was critically elevated at 81 μmol/litre (reference range 7–35 μmol/litre). A chest, abdomen and pelvis CT scan showed a cirrhotic liver with portal vein cavernous transformation, spontaneous, non-surgical hepatorenal shunt and varices. In addition, bilateral mild hilar lymphadenopathy (LAD) and a single, less than 2-mm right lower lobe pulmonary nodule were noted, which were radiologically considered to be consistent with sarcoidosis. Unfortunately, the pulmonary nodule was not amenable to biopsy thus precluding further confirmation of diagnosis. However, a liver biopsy was obtained to evaluate the cause of cirrhosis.
The liver biopsy showed hepatoportal sclerosis with proliferative vasculopathy (fig 4) and rare non-caseating granulomas (fig 5), which were negative for fungi by histochemical stains. In addition, there was dense fibrous tissue surrounding vascular structures. A panel for viral hepatitis was negative, and she did not abuse alcohol.
Figure 4.
Proliferative vasculopathy with reactive atypia of surrounding hepatocytes; arrow with tail denotes portal vein and arrow without tail is hepatic artery (H&E stain; original magnification ×20).
Figure 5.
Non-caseating granuloma with multinucleated giant cell in liver parenchyma (H&E stain; original magnification ×20).
However, the fact that there was extensive encasement of the brain with necrotising granulomas was not missed. Cerebrospinal fluid (CSF) was obtained and again cultures for AFB and fungi were sent, which were negative. In addition, a venereal disease research laboratory test performed on the CSF was also negative. CSF cytology was performed; rare macrophages and lymphocytes were noted, but it was negative for malignant cells.
Given the histology of her brain lesion, granulomas on liver biopsy and CT findings of the thorax, she was thought to be experiencing systemic sarcoidosis, and her brain lesion was best characterised as dural-based NS-NSG. However, given the extensive fibrosis seen in her dural-based masses as well as liver, multifocal fibrosclerosis, a rare idiopathic disease that involves multiple systems, was initially considered, but the presence of granulomas, hilar LAD and pulmonary nodule favoured the diagnosis of sarcoidosis. Although the patient’s NS was impressive, the majority of her altered mental status on follow-up was assessed to be due to hepatic encephalopathy as she responded rapidly to lactulose enemas.
DIFFERENTIAL DIAGNOSIS
The differential diagnosis of necrotising granulomas in the CNS is broad and includes numerous infectious agents such as mycobacteriosis, histoplasmosis, aspergillosis and cryptococcosis, as well as non-infectious causes including WG, foreign body granulomatosis, idiopathic pachymeningitis with granulomas formation and NS-NSG. Accordingly the management strategies vary greatly and range from an intensive antimicrobial regimen to immunosuppressants. Therefore, it is imperative to fully work-up any such case so that the appropriate therapy will be initiated.
Based upon the histopathology of the resected dural-based mass lesion, the two main differential diagnostic considerations were a mycobacterial infection (likely tuberculosis) and sarcoidosis. However, negative special stains, cultures for fungi and AFB on the lesional tissue, CSF studies and Mantoux test nearly ruled out an infectious aetiology. Tuberculoma mimicking an en plaque meningioma with a negative laboratory work-up for tuberculosis has been described in the literature.1 In this reported case, the resection revealed confluent necrotising and non-necrotising granulomas, and had a negative work-up for tuberculosis similar to our patient.
However, the patient had immigrated from India 5 years prior, his sister had a positive Mantoux test, and he personally responded to a trial of antituberculosis drugs. Our patient did not have any of the high risk factors for tuberculosis such as immunodeficiency, travel from an endemic area, or contact with infected individuals.
Rare differential diagnoses of NS include WG. CNS involvement has been reported in 2% to 8% of patients with WG, although intracranial granulomas are rare.2,3 Our patient did not meet the clinical or pathological criteria for WG. For instance, she did not have a history of sinusitis, nasal deformity, or oral ulcers; her renal function and anti-neutrophil cytoplasmic antibody screen were normal, and there was an absence of vasculitis on microscopic examination. Foreign body granulomatosis (such as berryliosis) was unlikely as no polarisable material was seen on histopathology.2 Finally, idiopathic pachymeningitis with granuloma formation is rare and requires excluding any underlying aetiology.4
TREATMENT
The patient’s original treatment plan consisted of a left frontotemporal craniotomy for resection of her dural-based mass. However, it was modified to a limited debulking once her intraoperative diagnosis was received. Concerning her hepatic encephalopathy, lactulose enemas were initiated. They were prescribed at 30 ml three times a day but were increased to 45 ml four times a day for better control of her ammonia levels. A trial of prednisone was implemented for the treatment of her NS-NSG with an initial dose of 60 mg and then tapered to a maintenance dose of 20 mg.
OUTCOME AND FOLLOW-UP
Currently, she is being followed on an outpatient basis. It has been 7 months since her original diagnosis of NS-NSG, and she continues to take a low dose of prednisone. At her most recent appointment, she was seizure free and noted to have a normal physical examination, except mild disinhibition. Furthermore, her ammonia level was normal.
DISCUSSION
Sarcoidosis is a complex multisystem disease that has a predilection for the lungs, hilar lymph nodes, eyes and skin, although it can affect any organ in the body including the CNS.5,6
The hallmark of sarcoidosis is non-caseating granulomas, often described as “tight” granulomas, in the absence of other aetiologies. It has been described in patients of all ages (usually under 50 years), sexes (more in women) and races. However, AAs are affected more often and tend to have a more severe and chronic disease, likely related to genetic associations within the human leukocyte antigen region.7 Abnormal laboratory findings such as pancytopoenia8 and elevated serological findings such as C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and angiotensin-converting enzyme (ACE) levels, are all non-specific.9 Interestingly, in our patient, a serum ACE level and high sensitivity CRP were normal; she was not on an ACE inhibitor at anytime. Although the lack of ACE level elevation is surprising given our patient’s extensive disease involvement, only half of patients with sarcoidosis have been described to have elevated ACE levels. Westergren ESR showed mild non-specific elevation at 55 mm/h (reference range 0–29 mm/h). Lymphopoenia was noted but no bone marrow biopsy was performed for evaluation of extrapulmonary sarcoid involvement. However, diagnosis of sarcoidosis is firm when chest radiographic evidence is accompanied by compatible clinical features and non-caseating granulomas on biopsy, with all other causes of granulomas ruled out.6 Despite its recognition as a specific disease more than 100 years ago, the cause of sarcoidosis is still unknown;2 possible contributors include a mycobacterial infection,6 nanoparticles6,10 and genetic factors.6,7
About 10% of all patients with sarcoidosis present with neurological symptoms commonly referred to as neurosarcoidosis.11,12 The most common problems, listed in decreasing order of frequency, are cranial nerve palsies, headache, ataxia, cognitive dysfunction, weakness and seizures.5,9,11–14 In addition to non-specific headaches, our patient had difficulty in word finding and visual disturbances. An elevated CSF ACE level is specific for NS but observed in only 33 to 58% of patient with confirmed NS (this level was normal in our patient).15 Early recognition of NS16 and an aggressive approach to treatment, with the use of corticosteroids and immunosuppressive agents such as azathioprine,12 have been advocated, as NS is one of the three main causes (along with pulmonary and cardiac involvement) of fatality in sarcoidosis, affecting about 5% of cases.2,6
Rare variants of sarcoidosis have been described and debated in the literature including necrotising sarcoid granulomatosis and nodular sarcoidosis.17–20 These are typically confined to lungs, but extrapulmonary cases have been described. Nodular sarcoidosis represents 1.6% to 4% of sarcoidosis cases; its hallmark is large radiographic pulmonary nodules (usually 1–5 cm), which are coalescent granulomas. Patients are typically younger with bilateral disease and hilar LAD; these patients often have a favourable prognosis. Nodular sarcoidosis is not typically associated with vasculitis or necrosis, and if these are present, a diagnosis of NSG is usually preferred. NSG requires identification of sarcoid-like granulomas, variable amounts of necrosis and granulomatous vasculitis. This granulomatosis vasculitis may consist of intraluminal granulomas causing destruction of portions of large vessel walls, destruction of walls of small and large vessels by an infiltration of inflammatory cells and histiocytes and/or compression of small vessels by granulomas (often this is accompanied with vessel occlusion secondary to intimal fibrosis).
Following the biopsy diagnosis of dural mass in our case, the patient was evaluated following the algorithm described in fig 1 by Costabel et al.16 The overall clinicopathological picture in our case best fits into NSG, albeit an example of its rather rare extrapulmonary presentation, namely NS-NSG. Only a handful of cases of NS-NSG have been reported in the literature.9,17,19,21,22 Furthermore, only four cases of NS-NSG presenting as meningeal en plaque/mass lesion, showing many clinicopathological similarities to our patient, have been reported in the literature (table 1).9,19,22
Table 1.
Comparison of reported patients with en plaque NS-NSG
| 1 | 2 | 3 | 4 | 5 | |
| Reference value | 5 | 11 | 20 | 20 | MCG |
| Age | 51 | 52 | 40 | 53 | 52 |
| Sex | M | F | F | M | F |
| Race | AA | AA | AA | W | AA |
| Symptoms | Visual complaints | HA, facial pain | Hearing loss, HA, tinnitus, dizziness | Leg pain, numbness, weakness, incontinence | HA, speech, visual complaints |
| No. of lesions | 2 | 1 | 1 | 1 | Multiple |
| Site | L frontal, R parietal | Temporal | R CP angle | Conus medullaris | Frontal |
| Size | Large | 1.8 cm | ? | ? | Large |
| PMH | Asthma | None | None | None | Thyroid dysfunction |
| CXR | Normal | Normal | Normal | ? | Normal |
| ACE levels | Normal | Normal | Normal | ? | Normal |
| ESR | Normal | Increased | ? | ? | Increased |
| CSF | ? | ? | >WBCs | >Protein, 0 WBCs | Normal |
| SS | Absent | Present | Absent | Absent | Present |
| Response to steroids | Yes | Yes | Yes | Yes | Yes |
| Follow-up time (months) | 18 | 24 | 12 | 9 | 7 |
| Active disease at follow-up | No | No | No | No | No |
AA, African–American; ACE, angiotensin-converting enzyme; CP, cerebellopontine; CXR, chest x ray; ESR, (Westergren) erythrocyte sedimentation rate; F, female; HA, headache; L, left; M, male; MCG, Medical College of Georgia; NS-NSG, neurosarcoidosis of the necrotising sarcoid granulomatosis variant; PMH, past medical history; R, right; SS, systemic sarcoidosis; W, white; WBC, white blood cells.
It is well known that although granulomas may resolve with little consequence, pulmonary fibrosis occurs in 20% to 25% of patients with sarcoidosis.2,6,18 Similar confluent fibrosis accompanying granulomas in the meninges likely resulted in these mass lesions. These cases showed excellent response to high dose steroid treatment. While there has been a favourable initial response in our patient, only time will tell if she will respond as well as the reported patients.
Other unusual findings about our patient were her history of hypothyroidism and current severe liver dysfunction. Anywhere from 1.3% to 54.5% of sarcoidosis patients have anti-thyroid antibodies often with a hypoechogenic pattern and decreased thyroid volume on ultrasound (particularly in women), thereby supporting an autoimmune aetiology.23 However, TSH levels were always within normal limits in our patient. An even more concerning and confusing finding was her recent hospital admission due to hepatic encephalopathy. Although 40% to 70% of patients with sarcoidosis have liver involvement,5,6,24 hepatic impairment secondary to sarcoidosis is rare.6 This rapid deterioration despite steroid treatment thus raised the concern that the granulomatosis might be secondary to an infective source. However, ultimately her altered mental status was reasonably assessed to be secondary to hepatic encephalopathy and not NS-NSG (or other potential cause of granulomatosis such as an infection), as she responded favourably to lactulose treatment.
In conclusion, sarcoidosis is a complex multisystem disease of unknown aetiology whose treatment is not standardised, and its prognosis cannot be easily predicted.2 However, our understanding of this varied disorder is growing. Herein a rather rare case of NS-NSG presenting as a dural-based en plaque mass, mimicking meningiomatosis cerebri, is presented with a discussion of the optimal approach to its diagnosis and management. NS-NSG should be considered in the differential diagnosis of intracranial masses, but should only be diagnosed after other infectious and non-infectious aetiologies have been judiciously ruled out.
LEARNING POINTS
The differential diagnosis of necrotising granulomas in the central nervous system is broad and includes chronic infections such as fungal organisms and tuberculosis, as well as non-infectious causes including Wegener granulomatosis, foreign body granulomatosis, idiopathic granulomatous pachymeningitis and neurosarcoidosis of the necrotising sarcoid granulomatosis variant (NS-NSG).
Neurosarcoidosis is primarily an intraparenchymal disease of the central nervous system that affects approximately 10% of patients with sarcoidosis.
NS-NSG is an exceedingly rare variant of sarcoidosis that can have atypical clinical presentations, including rarely intracranial mass lesions and non-alcoholic steatohepatitis (NASH) with hepatoportal sclerosis as in our case.
NSG requires identification of sarcoid-like granulomas, variable amounts of necrosis and granulomatous vasculitis. A thorough clinical and pathological work-up to exclude infectious and other non-infectious aetiologies is a prerequisite to the diagnosis of NS-NSG
Footnotes
Competing interests: None.
Patient consent: Patient/guardian consent was obtained for publication.
REFERENCES
- 1.Bauer J, Johnson RF, Levy JM, et al. Tuberculoma presenting as an en plaque meningioma. Case report. J Neurosurg 1996; 85: 685–8 [DOI] [PubMed] [Google Scholar]
- 2.Judson MA: Sarcoidosis: clinical presentation, diagnosis, and approach to treatment. Am J Med Sci 2008; 335: 26–33 [DOI] [PubMed] [Google Scholar]
- 3.Azuma N, Katada Y, Nishimura N, et al. A case of granuloma in the occipital lobe of a patient with Wegener’s granulomatosis. Mod Rheumatol 2008; 18: 411–15 [DOI] [PubMed] [Google Scholar]
- 4.Parney IF, Johnson ES, Allen PB. “Idiopathic” cranial hypertrophic pachymeningitis responsive to antituberculous therapy: case report. Neurosurgery 1997; 41: 965–71 [DOI] [PubMed] [Google Scholar]
- 5.Newman LS, Rose CS, Maier LA. Sarcoidosis. N Engl J Med 1997; 336: 1224–34 [DOI] [PubMed] [Google Scholar]
- 6.Iannuzzi MC, Rybicki BA, Teirstein AS. Sarcoidosis. N Engl J Med 2007; 357: 2153–65 [DOI] [PubMed] [Google Scholar]
- 7.Grunewald J. Genetics of sarcoidosis. Curr Opin Pulm Med 2008; 14: 434–9 [DOI] [PubMed] [Google Scholar]
- 8.Yanardağ H, Pamuk GE, Karayel T, et al. Bone marrow involvement in sarcoidosis: an analysis of 50 bone marrow samples. Haematologia (Budap) 2002; 32: 419–25 [PubMed] [Google Scholar]
- 9.Tobias S, Prayson RA, Lee JH. Necrotizing neurosarcoidosis of the cranial base resembling an en plaque sphenoid wing meningioma: case report. Neurosurgery 2002; 51: 1290–4 [DOI] [PubMed] [Google Scholar]
- 10.Heffner DK. The cause of sarcoidosis: the centurial enigma solved. Ann Diagn Pathol 2007; 11: 142–52 [DOI] [PubMed] [Google Scholar]
- 11.Joseph FG, Scolding NJ. Sarcoidosis of the nervous system. Pract Neurol 2007; 7: 234–44 [DOI] [PubMed] [Google Scholar]
- 12.Scott TF, Yandora K, Valeri A, et al. Aggressive therapy for neurosarcoidosis: long-term follow-up of 48 treated patients. Arch Neurol 2007; 64: 691–6 [DOI] [PubMed] [Google Scholar]
- 13.Zajicek JP, Scolding NJ, Foster O, et al. Central nervous system sarcoidosis - diagnosis and management. QJM 1999; 92: 103–17 [DOI] [PubMed] [Google Scholar]
- 14.Koczman JJ, Rouleau J, Gaunt M, et al. Neuro-ophthalmic sarcoidosis: the University of Iowa experience. Semin Ophthalmol 2008; 23: 157–68 [DOI] [PubMed] [Google Scholar]
- 15.Westhout FD, Linskey ME. Obstructive hydrocephalus and progressive psychosis: rare presentations of neurosarcoidosis. Surg Neurol 2008; 69: 288–92 [DOI] [PubMed] [Google Scholar]
- 16.Costabel U, Ohshimo S, Guzman J. Diagnosis of sarcoidosis. Curr Opin Pulm Med 2008; 14: 455–61 [DOI] [PubMed] [Google Scholar]
- 17.Gal AA, Koss MN. The pathology of sarcoidosis. Curr Opin Pulm Med 2002; 8: 445–51 [DOI] [PubMed] [Google Scholar]
- 18.Ma Y, Gal A, Koss MN. The pathology of pulmonary sarcoidosis: update. Semin Diagn Pathol 2007; 24: 150–61 [DOI] [PubMed] [Google Scholar]
- 19.Strickland-Marmol LB, Fessler RG, Rojiani AM. Necrotizing sarcoid granulomatosis mimicking an intracranial neoplasm: clinicopathologic features and review of the literature. Mod Pathol 2000; 13: 909–13 [DOI] [PubMed] [Google Scholar]
- 20.Kay JM. Correspondence re: Strictland-Marmol LB, Fessler RG, Rojiani AM. Necrotizing sarcoid granulomatosis mimicking an intracranial neoplasm: clinicopathologic features and review of the literature. Mod Pathol 2001; 14: 533. [DOI] [PubMed] [Google Scholar]
- 21.Leiba H, Siatkowski RM, Culbertson WW, et al. Neurosarcoidosis presenting as an intracranial mass in childhood. J Neuroophthalmol 1996; 16: 269–73 [PubMed] [Google Scholar]
- 22.Markert JM, Powell K, Tubbs RS, et al. Necrotizing neurosarcoid: three cases with varying presentations. Clinical Neuropathology 2007; 26: 59–67 [DOI] [PubMed] [Google Scholar]
- 23.Antonelli A, Fazzi P, Fallahi P, et al. Prevalence of hypothyroidism and Graves disease in sarcoidosis. Chest 2006; 130: 526–32 [DOI] [PubMed] [Google Scholar]
- 24.Drebber U, Kasper HU, Ratering J, et al. Hepatic granulomas: histological and molecular pathological approach to differential diagnosis - a study of 442 cases. Liver Int 2008; 28: 828–34 [DOI] [PubMed] [Google Scholar]





