Round 17: Facial Swelling

By: Christopher V. Tehlirian, MD

Dr. Tehlirian has no significant financial interest or relationships to disclose.

Release Date: February 5, 2009
Expiration Date: January 1, 2011


  • Present instructive medical cases
  • Teach relevant clinical data
  • Discuss treatment and management of those cases

Case 1:

A 19-year-old male presented with left facial swelling that began in the summer of 2001 at age 13. He woke up one morning and his brother pointed out that the left side of his face was swollen, particularly at the angle of the jaw. There was no pain associated with the swelling at that time. Over a period of months, his face continued to swell. He developed mild facial pruritis and pain associated with chewing. The swelling would slightly increase when he was in the shower. He went to multiple ear, nose, and throat physicians in 2002 and 2003, resulting in two fine needle aspirations with normal tissue diagnoses.

In June 2003, resection was done on some lymphoid tissue around the salivary gland. They could not get the whole thing out because of his left 7th nerve. Not wanting to leave him with a palsy, part of it was left in. There was a significant decrease in the size of what he had there. It remained decreased in size until September 2005. It started to enlarge on the left side, and swelling was noticed on the right side as well. He began to develop pruritis, particularly over the face, and mild dry mouth. There were no other complaints; he was eating well and feeling fine.

Medical, Family, and Social History

He had no past medical history. He was not taking any medications. He has no allergies. His father died in his 50s. He is of Arabic background. His mother is Filipino, has hypertension. There are no cancers or autoimmune diseases in the family. The patient has multiple siblings and half-siblings that are healthy with the exception of a few who have diabetes mellitus. He lives in the United Arab Emirates, and is studying engineering at college. He does not drink; he smokes about 3 to 4 cigarettes a day.

Physical Exam

Physical Exam Photo

On exam, he was mildly obese. His head and neck examination showed no lacrimal gland enlargement. His salivary pooling was fine. His oropharynx was clear. His teeth looked fine; nothing that would look like an abscess. ANCA was negative. He had no fever and did not have asthma or seasonal allergies. He was not taking any medications.

He had a mass over the left parotid that was about 7 x 6 cm. There was prominence of the submandibular glands, and no lymphadenopathy felt further down on the neck. There was a right pre-auricular enlargement that measured 5 x 4 cm. Otherwise, his cranial nerves were intact. His lung and heart exams were normal. The abdomen was normal. His liver seemed to be normal size. There were no other noted organomegalies. His spleen was not palpable. He had no rash, or history of one. Neurologically, he was fine. CT scans of the chest, abdomen, and pelvis showed he had some adenopathy around the left jugular-digastric area of his neck, but chest, abdomen and pelvis were clean.

ct exam

Essentially, he had a mild leukocytosis. The eosinophil count is elevated; absolute count is 1,200. His sedimentation rate is a little bit elevated; at 25 it is abnormal. His AGE is rather elevated at 6,000. Otherwise, the rest of his blood work was within reason. The same swelling of both parotid glands is show on the CT scan. There is some heterogeneity within the enlargement of the parotid gland.

Laboratory Findings


Na 145, K ESR 28 mm/hr
BUN 12, SCr 0.7 CRP 0.9 mg/dL
Ca 10.1 ANA negative
TP 7.9, Alb 4.4 SPEP normal
Tbili 0.4 IgG 1480, IgM 294, IgA 124
AST 40, ALT 37 IgE 6269 kU/L
Alk phos 97 Flow cytometry of blood negative
WBC 11.9K (eos 1280) HIV negative
HCT 44 UA normal
Plts 404 PPD negative

Tissue Analysis

Because of the parotid gland involvement and other lab findings, it was very difficult to figure out what this was without looking at the tissue.

(A) The parotid gland. Fat and secretory cells with huge lymphocitic infiltrate around them. Some neo-vascularization, not very prominent.

(B) Similarly, this is a germinal center within the salivary gland, which is supposed to be all red because it is all eosinophils. In fact, within the germinal center, it is just packed with lymphocytes, plasma cells, and eosinophils.

(C) A closeup image of eosinophils with lymphocytes surrounding them. There are no Reed-Sternberg cells or Hodgkin’s lymphoma.

(D) Lymphocytes and eosinophils next to each other with some laying down of collagen, fibrosis, and sclerosis which is a separate part of the lymph node within the salivary gland.

Discussion – Analysis (Kimura’s disease)

The tissue analysis is important, because it shows that this was Kimura’s disease. Kimm and Szeto first described Kimura’s disease in China in 1937. Kimura described the same thing a decade later, but added in the vascular aspects of the disease. The etiology is not known. It is more common in Asians, as shown in most case reports. There is a male to female ratio of 3-5:1. Kimura’s usually manifests with painless cervico-facial enlargement—subcutaneous masses with eosinophilia—which is an eosinophil-driven process. The IgE is markedly elevated (as in the above case), and it does not seem to be related to allergy. It is not known what drives the IgE elevation.

The prognosis is excellent; Kimura’s is benign and never becomes malignant. It usually affects young adults; the median age is around 28 years, and about one-third are recurrent. There is literature of descriptions of nephrotic syndrome associated with Kimura’s, particularly membranous arthritis. There is also misangioproliferative or other types of kidney disease with nephrotic syndrome, but membranous is the most commonly described.

Kimura’s disease is similar in appearance to angiolymphoid hyperplasia with eosinophilia, but there are differences in the histopathology. Up until about 1979, physicians were calling Kimura’s and angiolymphoid hyperplasia the same thing, or using one or the other name interchangeably. Kimura’s is more of an eosinophil- and IgE-mediated invasion of the lymph nodes within the salivary glands or cervical facial area with nodules. Angiolymphoid hyperplasia, on the other hand, is more of a very low-grade angiosarcoma. The angiosarcoma is associated with eosinophilia. Therefore, both Kimura’s and angiolymphoid hyperplasia have eosinophilia. The angiolymphoid hyperplasia is more of a blood vessel abnormality with proliferation with the tissue.
Angiolymphoid Hyperplasia with Eosinophilia vs. Kimura’s disease

Angiolymphoid Hyperplasia
with Eosinophilia
Kimura’s disease
Clinical features
Sex More often women Male predominance
Age Young to middle age Young adulthood
Race Occurs in all races More common in Asians
Presentation Nodules or erythematous papules Discrete nodules or localized swelling
Size Average 1cm Average 3cm
Location Usually head and neck Usually head and neck
Lymphadenopathy Uncommon Common
Blood eosinophilia In 20% of cases Almost invariably present
Histopathological features
Depth Dermis, subcutaneous Subcutaneous, muscle
Often circumscribed Usually noncircumscribed
Vessels Florid vascular proliferation Some degree of vascular proliferation
Endothelium Cuboidal to dome-shaped Flat to low
Inflammation Sparse to heavy infiltrate of lymphocytes and plasma cells Abundant lymphocytes and plasma cells
Lymphoid follicles May be present Always found
Eosinophils Sparse to abundant
Eosinophilic abscess rare
Moderate to abundant
Eosinophilic abscess present
Sclerosis Not a prominent feature Significant at all stages

Angiolymphoid Hyperplasia with Eosinophilia is treated mostly with corticosteroids and is always responsive to surgical resection. Therefore, with either of these two diseases the first step is usually surgical resection with or without intralesional or oral corticosteroids. There was a 2005 paper about Zyrtec being used in Kimura’s as a kind of a medication There are other small numbers of case reports using indomethacin, pentoxyfilline, and retinoids. The usual treatment is corticosteroids followed by cyclosporine (if the corticosteroids are not able to be weaned or if it is not as responsive). There are several papers outlining the use of corticosteroids and cyclosporine together. Radiation therapy is used as a final result if none of the previous therapies have been useful. There is a paper on radiation therapy by Cheng that looked at about 18 patients with Kimura’s. About 65% that were recalcitrant to other therapies responded to radiation therapy.

Case 2:

A 35-year-old man noticed that the submandibular area of his face started to swell after workouts in late Spring 2006. There was also some swelling in the hands. It lasted for a few moments, and then dissipated. In June 2006, he noted a few splinter hemorrhages, it is not clear whether he had traumatized perimadibular areas to cause this. In July 2006 facial swelling, unrelated to physical activity, started to worsen. His wife noticed that he was snoring more at night.

He went to his primary care physician, and found that his white blood cell count was 10,000, and that he had 23% eosinophils. He was sent to the allergist, who was worried about an eosinophilic leukemia, so he was then sent to an oncologist, who did a bone marrow in August 2006. The bone marrow was normal: 70% cellularity. The oncologist prescribed imatinib for about 5 days to drop his eosinophil count, which had been increasing over time. However, over the five days of use of imatinib his eosinophil count remained about 45–42%.

In September 2006, he went to a local ER when he felt that his throat was closing up and had severe difficulty breathing. They gave him IV solumedrol and sent him home with medrol. After two or three days the swelling in his face had almost dissipated. He noticed that he started urinating quite a bit, and he lost about 7-10 lbs overall. He tapered the medrol over the following 5 days, and over a 1-week-period after the taper, he did well. A week after stopping the medrol, he started to have similar symptoms recur. He was placed back on solumedrol, and then slow tapered.


  • Uulcerative colitis, diagnosed at age 14
  • Colectomy in 1994, complication during the surgery resulted in the removal of the spleen
  • History of thrombocytosis, which had been present since his splenectomy in 1994
  • Diagnosed with primary sclerosing cholangitis, for which he had been treated
  • Seasonal allergies from age 15, and sinusitis in the past
  • Mild, persistent asthma: one emergency room visit, necessitating a nebulizer treatment, never intubated
  • Uvulectomy in 2003.
  • Diagnosed with obstructive sleep apnea in 2007 (with chronic swelling he had a small obstruction in August 2006)
  • history of hernial repair. Sulfa causes angioedema.

His active medications were: Medrol (8 mg), Ursodiol for his primary sclerosing cholangitis, Claritin for allergies, Advair for asthma, Rhinocort as needed. Singulair was stopped for the workup of his eosinophil count; but he had been taking 10 mg a day.

His father is Persian; his mother is Irish-English, 58 years old, with a history of thrombocytosis, but no other problems with angioedema. No one else in the family has any problems with thrombocytosis or angiodema. He is an employment lawyer in California, married with children, and did not use tobacco or alcohol.

Physical Exam

On exam, his appearance was almost normal because he was on steroids at that point. Other than being rather cushingoid; he had no adenopathy, no enlargement of his face. His lung and heart exam was normal. His liver was normal size. On exam, he had no synovitis, but he had swelling in his hands in the past. What he described was on the flexor surfaces of his hand, particularly with the second and fifth proximal interphalangeal joints. He had little nodules that were more prominent in the past. They were not tender, but he had an uncomfortable sensation when he was grabbing things due to the nodules. Otherwise, neurologically he was fine. Below is what he looks like normally (right). and when he swells up (left).

Physical Exam - Case 2

He saw an orthopedic physician for the swelling or little nodules in his hands. MRI showed possible fibroma, no synovitis, and some questionable soft tissue swelling. The bone was fine. He had a CT scan of his neck by his primary care physician showing bilateral ethmoid and maxillary sinusitis, and mild soft tissue swelling. His facial swelling was slowly worsening; in fact, it was waxing and waning. Even before the steroids, it seemed to last for about 7 days or so, and then almost get back to normal completely, and then come back a few days later. It is strange in that respect.

He had a CT scan of the chest and abdomen that showed a small bowel obstruction. He had a lot of lymphadenopathy that was all subcentimeter. He ended up having an episode of hemoptysis where he coughed up a small amount of blood. While they were working him up for this eosinophilia, they also did a bronchoscopy and BAL. The BAL was normal. They did a transbronchial biopsy, which showed chronic bronchitis. There were no mild inflammatory cells, eosinophils, or granuloma on the biopsy of the lung. The bone marrow that was done in August 2006 showed 70% cellularity. His BCR-ABL was negative, and otherwise normal trilineage and increased eosinophils. He had pulmonary function tests that were normal. He had a blood PCR that was positive, and then the bone marrow was negative.

CBC’s over time

6/15/06 09/07/06 09/16/06 09/20/06 9/28/06
WBC (eosinophils) 12.0 (14%) 14.5 (42%) 30.1 (57%) 6.2 (29%) 9.0 (6.5%)
HCT 40 48 46 45.6 41
Plts 541 765 650 554 614

His symptoms started somewhere in May. Blood work from June to September showed that his white blood cell count wa going up and his percent of eosinophils was also going up. His hematocrit was stable. His platelets had been elevated since 1994 when he had the splenectomy. In addition, he had a bone marrow in 2000 that was normal. He had some mild eosinophil. He also was checked for parasites. The paragonimus IgG was positive. Strongyloides were negative. However, it is not clear what to do with this since he did not have any symptoms, did not grow anything from the stool. Blood work showed that his ANCAs were negative and SPEP was normal. His IgM, IgE, and IgG were normal. UA was normal. His rheumatoid factor was elevated at 158. His IgE was mildly elevated at 203. His inflammatory markers were mildly elevated.

His steroids were started on 9/16/06, Medrol (IV followed by PO). His white blood cell count went down, there were no nodules. He had a diffused adenopathy that is all less than 1 cm on his chest, abdomen, and pelvis. He had no history of cardiac disease or CHF. He felt better on steroids because the swelling was pretty uncomfortable.

Differential Diagnosis/Discussion–Episodic angioedema with eosinophilia (Gleich Syndrome)

Episodic angioedema with eosinophilia was first described by Dr. Gerald Gleich in 1984. This is thought to be an eosinophilic syndrome that is cyclical in nature, and—compared to the other hypereosinophilic syndromes—has a good prognosis. There is no organ involvement; there is no eosinophilic gastroenteritis. There is no endomyocardial fibroelastosis, no heart problems, and no eosinophilic pneumonia, so that is how this is different.

It is exceedingly rare, the symptoms and signs are:

  • Cyclical, non-allergic angioedema of the face, neck and extremities, and less likely trunk,
  • Urticaria,
  • Weight gain up to 14%, and then with steroids diurese,
  • Fever up to 38.5 °C,
  • Leukocytosis up to 50,000,
  • Eosinophilia up to 80%, with infiltration and degranulation in dermis
  • Elevated IgM, particularly during a flare.

Butterfield, et al shows a 22-day cycle. The attack happens around day 10. The white counts and eosinophils slowly start trending up. The interleukin-5 turned up before the eosinophils. Then with prednisone, the weight drops and the urine output picks up. It is not clear what the mechanism of this is; however, it is thought to be a TH-2 driven process.


In particular, there is a loss of equilibrium between TH-1 and TH-2, and for whatever reason, with these cycles, everything flips the TH-2, interleukin-4, 5, and 13, particularly interleukin-5. It started to get ramped up as well as interferon-γ, which precedes the attack. Then, following the surge of interleukin-5, white count goes up. They start activating eosinophils, which in turn release major basic protein. This activates basogenic edema and weight retention. With prednisone this is all shut off and they start diuresing, and they can lose the weight they had gained.

The cause is unknown, at least what triggers it. A group—Lassalle et al—in France that looked at anti-endothelial cell antibodies and whether or not these antibodies actually attach to the eosinophils, causing them to degranulate and release major basic protein, and then cause episodic angioedema with eosinophilia.
Incidence is unknown since there have only been a small number of cases described. It affects men and women; can be from ages 2-40 years. Prognosis is, just like Kimura’s, very good as there is no organ involvement with the eosinophilia.


Episodic Angioedema (Gleich’s syndrome) Non-episodic Angioedema
Younger age (16yo) Older (20s and 30s)
Female predominance (1:2) Females only
More in head/neck More in extremities
Recurrent Never recurrent
IgM elevated, possible IgE Normal IgM
May be recalcitrant to steroids Always responds to Steroids
Can experience arthralgia
No urticaria

Treatment and management is similar to other eosinophilic syndromes. Corticosteroids are the main stay. Hydroxyurea can be used; in fact, if corticosteroids are not viable, interferon-α and interleukin-5 inhibitors have become more and more popular. Non-episodic angioedema is treated with just steroids. Gleich’s syndrome tends to be more recurrent, so interferon-α can be used. This is actually used by Dr. Gleich, and so is interleukin-5.
The man in case 2 was sent to Dr. Gleich in Utah, and he believes that this gentleman has Episodic angioedema with eosinophilia; it is just that he has not had many episodes of it. The cryopyrin and something turned it on and turned it off. This is exactly the same where in effect you have almost similar two different types of clinical findings of disease. One is kind of on and stays on; one is on-off, on-off. Then what is the difference and what is driving that whole cascade of IgE and IL-5 is unclear.


In summary, these are two cases of interesting ENT manifestations of eosinophilic syndromes: The differentiation of Kimura’s and ALHE vs. episodic and non-episodic allergic angioedema. The current steroid-based management is similar for all of these eosinophilic syndromes; but interferon-α, and interleukin-5 are becoming more and more common with all of them.


  1. Mandel L, Surattanont F. Bilateral parotid swelling. Oral Medicine. 2002:93;221–237.
  2. Gumbs al. Kimura’s disease. A case report and literature review. Journal of Surg Onc. 1999;70:190–193.
  3. Klion A. et al. Approaches to the treatment of hypereosinophilic syndromes. J Allergy Clin Immunol.  2006:117;1292–1302.
  4. Seregard S. Angiolymphoid hyperplasia should not be confused with Kimura’s disease. Acta Ophthalmol Scand. 2001;79:91–3.
  5. Yamada A. et al. Membranous glomerular nephritis associated with eosinophilic lymphofolliculosis of the skin (Kimura’s disease). Clin Nephrol 1982:18;211-215
  6. Altman et al. Angiolymphoid hyperplasia with eosinophilia and nephrotic syndrome. Cutis.  1995:56;334–336.
  7. Wells and Whimster. Subcutaneous angiolymphoid hyperplasia with eosinophilia. Br J Dermatology. 1969:81;1–15.
  8. Ahn H et al. A clinicopathologic study of Kimura’s disease and Epitheloid Hemangioma. Yonsei Med J.  1990:31;205–211.
  9. Urabe et al. Epitheloid hemangioma versus Kimura’s disease. A clinicopathologic sudy. Am J Surg Pathol. 1987:11;758–766.
  10. Chen H, et al. Kimura’s disease. A clinicopathologic study of 21 cases. Am J Surg Pathol. 2004:28;505–153.
  11. Sato S, et al. Combined treatment of steroids and cyclosporine in Kimura disease. Pediatrics. 2006;118:e921–e923.
  12. Chang AR, et al. Outcomes of Kimura’s disease after radiotherapy or nonradiotherapeutic treatment modalities. Int J Radiat Oncol Biol Phys. 2006;5:1233–1239.
  13. Ben-Chetrit E, et al. Cetirizine: An effective agent in Kimura’s disease. Arthritis Rheum. 2005;53:117–118.
  14. Braun-Falco M, et al. Angiolymphoid hyperplasia with eosinophilia treated with anti-interleukin-5 antibody (mepolizumab). Br J Dermatol. 2004;151:1103–1104.
  15. Hongcharu W, et al. Kimura’s disease with oral ulcers: response to pentoxifylline. J Am Acad Dermatol. 2000;43:905–907.
  16. Butterfield JH, Gleich GJ. Interferon-alpha treatment of six patients with the idiopathic hypereosinophilic syndrome. Ann Intern Med. 1994;121:648–563.
  17. Butterfield JH, et al. Elevated serum levels of interleukin-5 in patients with the syndrome of episodic angioedema and eosinophilia. Blood. 1992;79:688–692.
  18. Lassalle P, et al. Presence of antibodies against endothelial cells in the sera of patients with episodic angioedema and hypereosinophilia. Clin Exp Immunol. 1990;82:38–43.
  19. Lorraine JK. Successful pregnancy in a woman with cyclic angioedema and eosinophilia. Ann Allergy Asthma Immunol. 1996;77:497–499.
  20. Gleich GJ, et al. Episodic angioedema associated with eosinophilia. N Engl J Med. 1984;310:1621–1626.

Updated: March 21, 2012

Arthritis Center

About Arthritis Center

Founded in 1998, the Arthritis Center at Johns Hopkins is dedicated to providing quality education to patients and healthcare providers alike.