Two Cases with Isolated Splenic Sarcoidosis Diagnosed by an Ultrasound-guided Fine-needle Aspiration Biopsy

Abstract

Asymptomatic splenic nodules were detected incidentally in two middle-aged women at an annual checkup. They showed no abnormalities on laboratory tests, but imaging studies revealed splenic nodules. No other localized lesions were found. Splenic nodules were hypoechoic on ultrasonography (US), hypovascular on contrast-enhanced computed tomography, and showed a low intensity on T2-weighted magnetic resonance imaging. We performed US-guided percutaneous aspiration biopsies using 21-gauge needles without complications, including bleeding. Pathological specimens showed noncaseating granulomas, so both patients were diagnosed with isolated splenic sarcoidosis. A US-guided fine-needle aspiration biopsy is a safe and useful method for diagnosing splenic nodules.

Introduction

Sarcoidosis is an idiopathic granulomatous disease characterized by the presence of systemic noncaseating granulomatous lesions (1). Although sarcoidosis can affect multiple parts of the body, most cases involve the pulmonary system, including the hilar lymph nodes (2).

The most important diagnostic criterion is histology. However, isolated organ sarcoidosis outside of the thoracic area is atypical, and its diagnosis can be difficult because the imaging findings are non-specific, and pathological examinations are often challenging.

We herein report two cases of isolated splenic sarcoidosis diagnosed pathologically by an ultrasound (US)-guided percutaneous fine-needle aspiration biopsy.

Case Reports

Case 1

A 43-year-old woman was admitted to our hospital for the evaluation and diagnosis of splenic nodules in June 2016. She had a history of excision of the extrahepatic bile duct and reconstruction for congenital choledochal cyst 10 years previously and had undergone magnetic resonance imaging (MRI) every year at a local hospital.

Although no localized lesion in the spleen had been noted in January 2014 (Fig. 1a), several nodules up to 1.0 cm in diameter had appeared in March 2015 (Fig. 1b). The nodules increased in size and number on images taken in January (Fig. 1c) and May (Fig. 1d) 2016.

Figure 1.

Case 1; fat suppressed T2 weighted MRI images at a) January 2014, b) March, 2015, c) January, 2016 and d) May, 2016.

A physical examination and laboratory findings were normal (Table 1). The splenic nodules were hypoechoic on US (Fig. 2a) and hypovascular on contrast-enhanced computed tomography (CT; Fig. 2b, c). They were also revealed to be isointense on T1- and showed low intensity on T2- and fat-suppressed T2-weighted MRI (Fig. 2d-f). Whole-body CT showed no localized lesions except for the splenic nodules.

Table 1.

Laboratory Findings on Admission of Case 1.

Hematology					ESR	4	mm/h
	WBC	6,700	/µL
	RBC	496×104	/µL	Serologyy
	Hb	14.3	g/dL		CRP	0.11	mg/dL
	Ht	43.6	%		IgG	1,180	mg/dL
	Plt	33.4×104	/µL		IgA	261	mg/dL
					IgM	250	mg/dL
Biochemistry					ANA	(-)
	TP	7.3	g/dL
	Alb	4.1	g/dL	Coagulation
	T-bil	0.5	mg/dL		PT-INR	0.98
	AST	20	IU/L		APTT	34.6	s
	ALT	20	IU/L
	LDH	197	IU/L	Others
	ALP	157	IU/L		sIL-2R	336	U/mL
	γGTP	17	IU/L		ACE	13.2	U/L
	BUN	15	mg/dL		T-SPOT	(-)
	Cre	0.73	mg/dL
	Na	138	Meq/L
	K	4.6	Meq/L
	Cl	104	Meq/L
	Ca	9.6	mg/dL

WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, Plt: platelet, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γGTP: gamma glutamiltranspeptidase, BUN: blood urea nitrogen, Cre: creatinine, Na: sodium, K: potassium, Cl: chloride, Ca: calcium, ESR: erythrocyte sedimentation rate, CRP: C-reactive protein, Ig: immunoglobulin, ANA: antinuclear antibody, PT: prothrombin time, sIL-2R: soluble interleukin-2 receptor, PT: prothrombin time, APTT: activated partial thromboplastin time, ACE: angiotensin-converting enzyme, T-SPOT: T-SPOT-TB

Figure 2.

Case 1; splenic nodules on admission. a) Ultrasonography showed a hypoechoic nodule, 11.2 mm (A) ×11.4 mm (B). Contrast enhanced CT at b) arterial phase and c) equilibrium phase. Plain MRI on d) T1, e) T2 and f) fat suppression T2 weighted images.

To make a pathological diagnosis, we performed a US-guided percutaneous aspiration biopsy of the splenic nodules using a 21-gauge needle with negative suction once without complications. Histological sections showed compact epithelioid cell granulomas containing Langhans-type multinucleated giant cells with small necrotic foci on hematoxylin and eosin staining (Fig. 3a, b) and no microorganisms, such as acid-fast bacilli or fungi, on Ziehl-Neelsen, Periodic acid-Schiff (PAS) and Grocott staining.

Figure 3.

Pathological specimens (Hematoxylin and Eosin staining) obtained by 21-guage needle aspiration biopsies, a) b) case 1 and c) d) case 2. They showed compact epithelioid cell granulomas. Langhans-type multinucleated giant cells were contained in case 1 (arrows), but not in case 2.

Echocardiography and an ophthalmological examination also showed no abnormalities. Therefore, we diagnosed her with isolated splenic sarcoidosis and observed her without medication. The splenic lesions grew slightly until November 2016 and then began decreasing gradually without new lesions.

Case 2

A 52-year-old woman presented with no symptoms, including weight loss, gastrointestinal symptoms, a fever or night sweats. She had been in good health and had no history of abnormalities at annual medical checkups, including abdominal US.

A 2.5-cm diameter solitary hypoechoic nodule was noted in the spleen by US at an annual medical checkup for the first time in 2016, and she subsequently visited our hospital in September 2016. A physical examination showed no specific findings. Laboratory test results revealed mild elevation of serum soluble interleukin-2 (IL2) receptor (Table 2). US showed a homogenous hypoechoic nodule with a clear margin in comparison to the surrounding parenchyma (Fig. 4a). The nodule showed isodensity on plain CT and appeared to be hypovascular in early- and late-phase contrast-enhanced CT (Fig. 4b-d). Whole-body CT and gallium scintigraphy revealed no localized lesions except for the splenic lesion (Fig. 4e). The splenic nodule was isointense on T1-weighted imaging and showed a high intensity with a low-intensity rim on T2- and fat-suppressed T2-weighted MRI (Fig. 4f-h). On gadolinium-enhanced MRI, the lesion was hypovascular in the early phase and gradually enhanced from the late phase to the equilibrium phase (Fig. 4j, k).

Table 2.

Laboratory Findings on Admission of Case 2.

Hematology					ESR	9	mm/h
	WBC	5,300	/µL
	RBC	480×104	/µL	Serologyy
	Hb	13.8	g/dL		CRP	0.04	mg/dL
	Ht	42.1	%		IgG	1,329	mg/dL
	Plt	28.2×104	/µL		IgA	263	mg/dL
					IgM	75	mg/dL
Biochemistry					ANA	(-)
	TP	7.5	g/dL
	Alb	4.4	g/dL	Coagulation
	T-bil	0.6	mg/dL		PT-INR	1.00
	AST	26	IU/L		APTT	30.1	s
	ALT	20	IU/L
	LDH	202	IU/L	Others
	ALP	155	IU/L		sIL-2R	555	U/mL
	γGTP	16	IU/L		ACE	6.4	U/L
	BUN	17	mg/dL		QFT	(-)
	Cre	0.68	mg/dL
	Na	140	Meq/L
	K	4.4	Meq/L
	Cl	104	Meq/L
	Ca	9.6	mg/dL

WBC: white blood cell, RBC: red blood cell, Hb: hemoglobin, Ht: hematocrit, Plt: platelet, TP: total protein, Alb: albumin, T-bil: total bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, γGTP: gamma glutamiltranspeptidase, BUN: blood urea nitrogen, Cre: creatinine, Na: sodium, K: potassium, Cl: chloride, Ca: calcium, ESR: erythrocyte sedimentation rate, CRP: C-reactive protein, Ig: immunoglobulin, ANA: antinuclear antibody, PT: prothrombin time, sIL-2R: soluble interleukin-2 receptor, PT: prothrombin time, APTT: activated partial thromboplastin time, ACE: angiotensin-converting enzyme, QFT: QuantiFERON-TB00

Figure 4.

Case 2; splenic nodule on admission. a) Ultrasonography, b) plain CT, contrast enhanced CT at c) early phase and d) late phase, and e) gallium scintigraphy. Plain MRI on f) T1, g) T2 and h) fat suppressed T2 weighted images, and gadolinium enhanced MRI at i) early phase, j) late phase and k) equilibrium phase.

We performed a US-guided percutaneous aspiration biopsy of the splenic nodule using a 21-gauge needle with negative suction once without complications, and as a result, we made a diagnosis of isolated splenic sarcoidosis (Fig. 3c, d) and observed her without medication. The serum level of soluble IL2 receptor returned to the normal range in 2018. The splenic solitary lesion had grown to 4.5 cm in diameter in 2018 and then started decreasing gradually, showing a diameter of 2.8 cm in 2021.

Discussion

The exact cause of sarcoidosis is still unknown, although it is suggested that genetic susceptibility, environmental factors and immune response to unidentified antigens contribute to disease development (2-5). The primary location of these lesions is the lung or the related lymph nodes in more than 90% of patients (6,7). Although extrapulmonary involvement was found in 30% of patients with the disease, isolated extrapulmonary manifestations of sarcoidosis occur in only 10% of cases (8,9). Splenic involvement is uncommon (5-10% of cases) and is complicated with pulmonary lesions in most sarcoidosis patients (2,10). Isolated splenic sarcoidosis is extremely rare.

We searched PubMed with the terms “isolated splenic sarcoidosis” and “primary splenic sarcoidosis” for articles published in English until December 31, 2021. We found 14 isolated splenic sarcoidosis cases and summarized their details in Table 3, including our own 2 cases (11-23). Isolated splenic sarcoidosis was more common in women than in men, and abdominal pain and weight loss were frequently identified as subjective symptoms. Serum levels of calcium and angiotensin-converting enzyme (ACE) were normal in many cases and mildly elevated in others. Splenic lesions were multiple in 13 cases. Unlike our two cases, the previous cases were all diagnosed by splenectomy.

Table 3.

Summary of the Reported Cases with Isolated Splenic Sarcoidosis.

No	Reference	Age/	Symptoms	Calucium	ACE	Number of	Diagnosis
		sex		(normal range)	(normal range)	splenic lesions
1	11	47/F	Nausea, epigastric pain	-	-	Multiple	Splenectomy
2	12	53/F	Abdominal pain	9.3 mg/dL (8.5-10.5)	45 U/L (8-52)	Single	Splenectomy
3	12	32/F	Abdominal pain	Normal	66 U/L	Single	Splenectomy
4	13	46/F	Back and leg pain	3.72 mmol/L	36 U/L	Multiple	Splenectomy
5	14	18/M	Night sweat	-	-	Multiple	Splenectomy
6	15	71/F	Cold sweat, palpitation	-	-	Single	Splenectomy
7	16	74/F	Weight loss, malaise, anorexia	3.64 mmol/L (2.0-2.6)	-	Multiple	Splenectomy
8	17	66/F	Anemia	-	-	Multiple	Splenectomy
9	18	37/F	Abdominal discomfort	-	-	Multiple	Splenectomy
10	19	29/F	none	-	-	Multiple	Splenectomy
11	20	65/M	Weight loss, headache	14.1 mg/dL (8.5-10.5)	8 U/L (9-67)	Multiple	Splenectomy
12	21	50/F	Weight loss, anorexia	-	-	Multiple	Splenectomy
13	22	40/F	Low grade fever, night sweat	-	-	Multiple	Splenectomy
14	23	76/F	None	-	-	Multiple	Splenectomy
15	Case 1	43/F	None	9.6 mg/dL (8.5-10.2)	13.2 U/L (8.3-21.4)	Multiple	Percutaneous biopsy
16	Case 2	52/F	None	9.6 mg/dL (8.5-10.2)	6.4 U/L (8.3-21.4)	Single	Percutaneous biopsy

ACE: angiotensin-converting enzyme

The serum level of soluble IL2 receptor was mildly elevated in our case 2. Soluble IL2 receptor is a representative marker of non-Hodgkin's lymphoma and is also well known to be elevated in sarcoidosis. Kita et al. compared the serum levels of soluble IL2 receptor in patients with sarcoidosis and non-Hodgkin's lymphoma (24), reporting level ranges of 195-3,750 U/mL in sarcoidosis and 240-62,300 U/mL in non-Hodgkin's lymphoma. The soluble IL2 receptor level in stage III/IV non-Hodgkin's lymphoma patients was significantly higher than in sarcoidosis and stage I/II non-Hodgkin's lymphoma patients. However, the serum levels of soluble IL2 receptor were not markedly different between sarcoidosis and stage I/II non-Hodgkin's lymphoma patients. Therefore, sarcoidosis cannot be distinguished from non-Hodgkin's lymphoma by soluble IL2 receptor levels alone.

The consensus recommendation for a definite diagnosis of sarcoidosis requires a demonstration of typical noncaseating granulomata in a pathological specimen. Imaging examinations and serum markers, such as ACE and calcium, are very useful, but the diagnosis of sarcoidosis (1) is difficult if the primary lesion is in a solitary organ and in an uncommon location, as was the case in our two patients (25). Isolated splenic sarcoidosis usually does not cause symptoms such as abdominal pain, a fever, malaise, and weight loss. Laboratory tests are also not usually helpful for the diagnosis of isolated splenic sarcoidosis (25). Ultrasonography typically shows splenic sarcoidosis as a hypoechoic lesion, like other splenic tumors. In addition, splenic sarcoidosis often shows different degrees of echogenicity (hypo, iso or hyper) according to the degree of fibrosis in the nodules (26). On MRI, splenic sarcoidosis shows a low signal in all sequences, including T2- and fat-suppressed T2-weighted images. On dynamic enhanced CT and MRI, splenic sarcoidosis is hypovascular in the early phase and gradually enhanced from the late to equilibrium phases. Gallium scintigraphy reveals it as a localized lesion. However, these image findings of splenic sarcoidosis are nonspecific, and it is difficult to distinguish splenic sarcoidosis from lymphoma, metastasis, angiosarcoma, hemangioma, hematoma, abscess, hamartoma and inflammatory pseudotumor (27). As a result, a pathological diagnosis is achieved in most patients by splenectomy. However, half of all cases with sarcoidosis resolve spontaneously within two years, and many of the remaining cases do so within five years (2,28-30). Furthermore, most cases of isolated splenic sarcoidosis resolve asymptomatically without treatment (2). Therefore, the accurate diagnosis of isolated splenic sarcoidosis is very important.

Percutaneous image-guided procedures of the spleen, including a biopsy, fluid aspiration and drainage, have not been widely performed historically because of the perceived increased risk of complications (31,32). The representative complication of a splenic biopsy is hemorrhaging. Such patients are managed with aggressive fluid resuscitation, blood transfusion and invasive procedures, including transcatheter arterial embolization and splenectomy. Other less-common complications are pneumothorax, pleural effusion and colonic injury. However, some reports suggest much lower rates of complications than 6.8% (33-35).

There are two methods of performing a percutaneous splenic biopsy: a core needle biopsy and a fine-needle aspiration biopsy. A core needle biopsy is usually performed using spring-loaded 14- to 20-gauge needles. A fine-needle aspiration biopsy is done using 20- to 23-gauge needles with negative suction (36). Compared to fine-needle aspiration, a core needle biopsy is slightly superior for making a diagnosis but inferior with regard to safety, as complication rates are higher with large-caliber core needles than with small-caliber aspiration needles. It was reported that complication rates of percutaneous splenic biopsies were 12% with a 14-gauge core needle and rare with 22-gauge aspiration needles (31,37). It is also reported that the rate of complications requiring medical intervention was 6.8% among 44 percutaneous image-guided splenic fine-needle aspiration biopsies (32). Therefore, fine-needle aspiration should be considered when performing a percutaneous biopsy of splenic lesions.

Singh et al. (36) recommended that laboratory criteria for the procedure be a platelet count of ≥50,000 /μL, prothrombin time (international normalized ratio) of ＜1.2 and activated partial thromboplastin time of 20-33 seconds. With regard to the accuracy of the diagnosis, the respective sensitivities of core needle and fine-needle aspiration biopsies were 88% and 80% (21). Therefore, fine-needle aspiration is superior in terms of safety and similar to a core needle biopsy in terms of diagnostic accuracy. We were able to safely diagnose the splenic nodules by the fine-needle aspiration procedure in our cases.

A percutaneous biopsy for splenic lesions has some limitations. If visualization of splenic lesions is difficult due to interposed lung, colon and ribs, percutaneous procedures are risky. Obesity, ascites and recent abdominal surgery are also risk factors for a percutaneous biopsy (38,39). In addition, lesions ＜3 mm in size or located in the splenic hilus are difficult to approach by a percutaneous biopsy (40). In such cases, splenectomy or an endoscopic approach are considered for diagnosing splenic lesions. However, splenectomy is invasive and has long-term immunological disadvantages. Endoscopic US-guided fine-needle aspiration (EUS-FNA) is useful and less invasive than splenectomy (40). Lisotti et al. conducted a comprehensive systematic review of EUS-FNA for splenic lesions and reported that the pathological diagnostic accuracy and adverse event rates were 88.2% and 4.7%, respectively (38).

In conclusion, we encountered two cases of isolated splenic sarcoidosis diagnosed by a US-guided 21-gauge needle aspiration biopsy. It has been difficult to diagnose isolated splenic sarcoidosis without surgical splenectomy. However, a US-guided fine-needle aspiration biopsy of splenic nodules is a simple, safe and useful technique that should be considered when a biopsy of splenic nodules is required.

The authors state that they have no Conflict of Interest (COI).