To our knowledge, mycotic infection of the subacromial bursa has not been
reported in the literature. We present the case of a patient with the classic
symptoms of impingement syndrome who was found to have a Candida infection of
the subacromial bursa. The patient was successfully treated with subacromial
decompression and acromioplasty to relieve the impingement as well as with
empirical antibiotic treatment for the infection. Our patient was informed
that data concerning the case would be submitted for publication.
A sixty-five-year-old, right-hand-dominant man was first seen because of
progressively increasing pain of more than four months' duration in the right
shoulder. He had pain at night, especially when sleeping on the affected side.
The patient also reported that the pain limited the range of motion, which
restricted his overhead function as well as his ability to reach behind his
back. He was barely able to tuck in the back of his shirt, and he had weakness
of the affected arm. The patient stated that he had injured the right shoulder
while lifting something at work ten months prior to being seen by a physician
and that the symptoms had become progressively worse. The patient had been
seen by another orthopaedist and was treated unsuccessfully with subacromial
injections three times over the course of two months. The subacromial
injection consisted of 40 mg of Depo-Medrol (methylprednisolone acetate), 4 mL
of lidocaine, and 4 mL of 0.25% Marcaine (bupivacaine). All injection
medication was withdrawn from single-use vials. The patient's response to each
injection was minimal, with some pain relief with motion initially but with a
rapid return to the preinjection state of discomfort. We first saw the patient
two months after the onset of symptoms. The medical history was important only
for a myocardial infarction sixteen years previously. He reported no history
of fevers or other systemic symptoms. The patient had routine preoperative
blood studies that showed a white blood-cell count of 5.4
cells/mm3, which remained within the normal range throughout the
course of treatment. He had no history of immunodeficiency, oral
corticosteroid use, or any other disease process that would explain an
opportunistic infection. His medications included lisinopril, simvastatin, and
rofecoxib. The patient stated that he did not smoke cigarettes, drink alcohol,
or use illicit drugs.
On the initial physical examination, no atrophy or asymmetry of the upper
extremities or the shoulder girdle musculature was noted. Upon palpation of
the right shoulder, no tenderness could be elicited. The active range of
motion of the right shoulder was 160° of forward flexion, 90° of
abduction, and 45° of external rotation. Internal rotation was to L5-S1.
All active and passive range of motion was limited by pain, especially at the
end points of the arc of motion. Passive motion was the same as active motion,
again limited by pain. The range of motion of the left shoulder was normal.
Both the Neer1 and
the Hawkins1
impingement signs were positive on the right. The supraspinatus,
infraspinatus, and subscapularis all had grade-4 (of 5) strength. The
neurovascular function was intact.
Plain radiographs of the right shoulder were unremarkable. Magnetic
resonance imaging of the right shoulder revealed lateral downsloping of the
acromion and mild thickening of the coracoacromial ligament with impingement
on the subacromial bursa. Subacromial bursal thickening and fluid accumulation
was also noted. Additional findings of note on the magnetic resonance imaging
scans were tendinosis of the subscapularis tendon with no evidence of a tear,
an attenuated and irregular long head of the biceps tendon, and glenohumeral
joint-fluid accumulation, which was most pronounced in the region of the
subscapularis recess and the biceps tendon sheath. All of these findings were
consistent with the diagnosis of impingement syndrome, and at that time there
was no evidence of an abscess or infection of the subacromial bursa.
The patient was treated conservatively with physical therapy for the next
few months. After one month, the symptoms were resolving, and strength as well
as range of motion also was improved. The patient stated at that time that he
felt "75% to 80% better." Therapy was continued for another month.
When he was seen at the two-month follow-up examination, the patient requested
surgery because he thought that the benefits of physical therapy had reached
their limit and he still felt discomfort. At that point, the patient had not
had any episodes of fever, increased pain, or tenderness of the right
shoulder. Nonoperative treatment was continued for a total of nine months
before a decision was made to proceed with surgery.
The patient then underwent arthroscopic subacromial decompression with
acromioplasty. Examination with the patient under anesthesia revealed the
range of motion to be identical to that at the time of the initial physical
examination (described above) with no evidence of loss of passive range of
motion or adhesive capsulitis. Through standard anterior and posterior
portals, the glenohumeral joint was visualized. Capsular inflammation was
noted superiorly as well as in the rotator interval. Attention was then
directed toward the subacromial space, where extensive proliferation of the
subacromial bursa and an inflammatory response were seen
(Fig. 1). Under direct
visualization, a lateral portal was made for better access to all aspects of
the subacromial space with the shaver. The impingement lesion involving the
coracoacromial ligament was visualized and taken down, and an enlarged
acromial spur was removed. The subacromial space was completely decompressed
with complete resection of the subacromial bursa.
Tissue specimens from the shaver, consisting of multiple pieces of
white-yellow, focally brown soft-tissue that measured in aggregate
approximately 5 × 4 × 1.5 cm in greatest dimension, were sent for
pathologic examination. Histologically, fragments of synovium, dense
collagenous tissue, bone, and skeletal muscle could be seen. The synovium
showed marked hypertrophy and hyperplasia with a chronic inflammatory
infiltrate in the subsynovial region. Numerous granulomas consisting of
foreign-body giant cells and epithelioid cells were seen throughout the
synovium. The giant cells in the granulomas had a c-shaped to centrally
located configuration of the nuclei (Fig.
2). Focal areas of infarction were also present. Scattered
throughout the tissue were numerous round to oval structures with a dark rim
and occasionally with clearing in the central region. Some of these structures
showed budding forms and were arranged singly as well as grouped in clusters,
perhaps representing some type of fungal organism. Special staining techniques
with use of Gomori methenamine silver stain, periodic acid-Schiff reaction,
periodic acid-Schiff reaction after treatment with diastase, and mucicarmine
confirmed the presence of round to oval structures that were consistent with
yeast cells (Fig. 3). Special
stains for bacteria and acid-fast bacilli were all negative. These morphologic
features were consistent with a granulomatous inflammation secondary to a
fungal infection, most likely the Candida species. No cultures were performed
to ascertain the exact organism in this case.
The patient was treated with a standard postoperative rehabilitation
protocol for impingement syndrome. In addition, he was managed by an
infectious disease physician and was started on an empirical dose of 200 mg of
voriconazole administered orally twice a day. He was maintained on this
antifungal medication for six weeks. At the one-year postoperative follow-up
visit, the patient stated that he had not had any episodes of fever, chills,
erythema, swelling, or drainage from the affected shoulder. Although he still
complained of a feeling of "residual achiness" in the right
shoulder, no point tenderness could be elicited at the time of the physical
examination. The active range of motion at that time was 140° of forward
flexion, 120° of abduction, and 90° of external rotation. The passive
range of motion was 160° of forward flexion and 130° of abduction. We
believe that the patient had an improvement of 70%, and we were not sure what
was preventing his complete recovery. Since the patient was still experiencing
persistent shoulder stiffness, his rehabilitation program in physical therapy
was continued.
Infection of the subacromial space has been rarely reported, with only a
few such cases in the literature, most of which had some component of
immunocompromise involved in the onset of the
infection2-6.
In the present case, our patient was healthy without a history of any disease
that would lead us to believe that he had increased susceptibility to an
opportunistic infection, and there was no evidence that he was
immunocompromised.
Septic bursitis is commonly seen in a more superficial bursa such as the
olecranon or the prepatellar bursae. These sites are easily inoculated
transcutaneously by means of minor trauma. Sporothrix
schenckii7,
Candida
lusitaniae8,
Mycobacterium
goodii9,
Paecilomyces
lilacinus10,
Mycobacterium
asiaticum11,
Anthopsis
deltoidea12,
Prototheca
zopfii13,
and Aspergillus
terreus14
are just a few of the many organisms that have been found to be the cause of
infection of the superficial bursae. In contrast, the location of the
subacromial bursa, deep to the deltoid muscle and the coracoacromial arch and
superficial to the rotator
cuff2,3,
leaves it relatively protected from transcutaneous inoculation. When infected,
deep bursal locations such as the subacromial bursa are more commonly infected
with Staphylococcal
species3.
Infection of the subacromial bursa is rare and typically has been related
to three important factors: (1) immunocompromise, (2) injection into the
subacromial space, and (3) hematogenous
spread2,3.
Co and Baer3
reported the case of a patient who abused intravenous drugs and presented with
a Staphylococcus aureus infection of the subacromial bursa, which was
diagnosed by aspiration of purulent fluid from the subacromial space. Upon
presentation, the patient had a fever, an elevated white blood-cell count, and
limited range of motion of the affected shoulder. The patient was treated
successfully with repeated bursal aspirations, intravenous nafcillin, and
physical therapy. At the six-month follow-up examination, he had a full range
of motion of the shoulder, and no bursal swelling was noted. Ward and
Eckardt4 reported on
four patients who were systemically ill or immunocompromised with subacromial
abscesses, which, when cultured, grew Staphylococcus aureus. All four
patients had pain, fever, and a decreased range of motion of the affected
shoulder, and all were successfully treated with open drainage of a gross
collection of pus from the subacromial bursa and antibiotic therapy. Hiemstra
et al.2 reported on
three patients who had a subacromial infection develop after a corticosteroid
injection for the treatment of subacromial impingement. Each patient presented
with pain, fever, and swelling shortly after the injection. In all of them,
grossly purulent material was found in the subacromial bursa at the time of
surgical drainage, and Staphylococcus aureus was cultured. Garcia
Mata et al.5
reported on an immunocompetent eleven-year-old child with no previous symptoms
who was found to have a subacromial bursal abscess. He was treated
successfully with open drainage and antibiotics.
We could not find a single report of an immunocompetent patient with a
fungal infection of the subacromial bursa. Furthermore, we know of only one
reported case of an immunocompromised patient with a Candida albicans
infection at this
site6. The patient
was managed with chronic corticosteroid therapy for the treatment of systemic
lupus erythematosus, and the complication developed after successful treatment
of a Staphylococcal infection of the infrapatellar bursa.
We cannot explain why our patient had only symptoms of impingement without
fever, swelling, or an elevated white blood-cell count. Although the argument
can be made that these findings are the result of specimen contamination, the
presence of granuloma formation as well as foreign-body giant cells and the
chronic inflammatory infiltrate in the subsynovial region indicate that this
infectious process was present for some time.
This is the first known episode of infection after a subacromial injection
in several hundred patients at our institution. We do not believe that it
warrants a change in our clinical management of patients with routine
impingement symptoms. Infection should always be considered in a patient with
shoulder pain, particularly following several subacromial injections.
?