Case 1. A nineteen-year-old right-hand-dominant competitive
gymnast and diver presented with symptoms of left-sided shoulder instability,
which had first developed when he was fifteen years of age and had been
treated conservatively for three years; he had no dislocations. Physical
examination demonstrated 2+ anterior load and shift, 2+ posterior load and
shift, a 2+ sulcus
sign7, positive
apprehension and relocation tests, and no signs of generalized ligamentous
laxity. A magnetic resonance imaging-arthrogram demonstrated a normal
glenohumeral joint and a patulous capsule
(Fig. 1). Despite nonoperative
management, including activity modification, physical therapy, and treatment
with anti-inflammatory medications, the instability pattern persisted and the
patient elected to proceed with arthroscopic surgery. The intraoperative
findings demonstrated a patulous capsular pouch. The treating surgeon believed
that thermal capsulorrhaphy was indicated on the basis of the preoperative
examination and the intraoperative findings and treated the capsule thermally
with the "zebra-striping" technique, which is designed to leave
normal strips of capsular tissue intertwined with the thermally altered
tissue. A monopolar radiofrequency device (Oratec ORA-50; Oratec
Interventions, Menlo Park, California) was used with the standard settings of
67°C and a power of 40 W. Irrigation flow was set on low during the
procedure, and the fluid was at room temperature. The chondral surfaces were
normal, and no thermal treatment of the chondral surfaces was performed.
The shoulder was placed in a sling immediately postoperatively, and a
progressive range-of-motion exercise program was begun at two weeks. By eight
weeks after surgery, the patient had regained a nearly full range of motion.
He had no pain and no instability and had normal function for six months,
during which time he returned to competitive diving. Slowly, however, left
shoulder pain, a decreased range of motion, and decreased function developed
and began to affect his diving performance. Radiographs
(Figs. 2-A and 2-B) and a
magnetic resonance image (Fig.
3) demonstrated complete obliteration of the joint space, and a
provisional diagnosis of chondrolysis was made. The pain became constant and
was poorly controlled with oral pain medication. Despite six months of
physical therapy following the diagnosis of chondrolysis, the pain,
dysfunction, and range of motion continued to worsen.
The patient was then referred to us for further evaluation and treatment.
On physical examination, the active range of motion of the left shoulder
consisted of forward elevation to 30°, external rotation to
—20°, and internal rotation to the L5 level. The passive forward
elevation was 60°, and abduction was 30°. The range of motion of the
contralateral, right shoulder consisted of 180° of forward elevation,
60° of external rotation at the side, 90° of external rotation in
90° of abduction, and internal rotation to the T7 level. The patient had
profound tenderness to palpation of the left shoulder and mild atrophy of the
left upper extremity. All major muscle groups had 5 of 5 strength. Sensation
and reflexes were all normal. The results of laboratory studies, including the
white blood-cell count, the measurement of erythrocyte sedimentation rate, and
the measurement of the level of C-reactive protein, were normal, and cultures
of specimens from the glenohumeral joint were negative.
Because of the age of the patient, a standard total shoulder replacement
was not performed as a result of concerns regarding polyethylene glenoid
loosening and the extent of bone resection necessary for a standard humeral
head replacement. Therefore, the patient elected to undergo a resurfacing
arthroplasty on the humeral side and a meniscal allograft resurfacing
procedure on the glenoid
side8. A standard
deltopectoral approach was used. Findings included a sclerotic glenoid without
articular cartilage, a denuded humeral head with gross destruction of the
normal osseous anatomy, and profound synovitis
(Fig. 4). A medium
hydroxyapatite-coated humeral head resurfacing prosthesis (Copeland shoulder
cementless surface replacement; Biomet, Warsaw, Indiana) was fitted, and a
lateral meniscal allograft glenoid resurfacing procedure was then performed.
The lateral meniscal allograft was obtained from the American Red Cross
Transplantation Services (Eatontown, New Jersey). The allograft was
fresh-frozen and treated with low-dose gamma irradiation (1.0 to 1.8 Mrad
[10,000 Gy to 18,000 Gy]) to reduce the bioburden. The patient underwent a
standard postoperative protocol to regain motion during the first six weeks.
Progressive active range-of-motion and strengthening exercises were instituted
at six weeks and were continued for six months. At the one-year follow-up
examination, the patient had active range of motion that consisted of 160°
of forward elevation, 50° of external rotation at the side, 90° of
external rotation in 90° of abduction, and internal rotation to the T10
level. He had no pain at the time of the most recent follow-up, and the most
recent radiograph demonstrated good preservation of the new glenohumeral joint
space (Fig. 5).
Case 2. A twenty-year-old right-hand-dominant man presented to
another institution with symptoms of anterior shoulder instability. He had a
history of four traumatic anterior dislocations of the left shoulder, all of
which required assisted reduction. The first dislocation occurred when the
patient was seventeen years of age, during a game of football. The physical
examination showed a positive apprehension sign in abduction and external
rotation as well as a positive relocation maneuver. Nonoperative management
(physical therapy, activity modification, and the use of anti-inflammatory
medication) failed, and the patient agreed to undergo surgery. The surgeon
elected to perform arthroscopic labral repair as well as thermal
capsulorrhaphy with use of a monopolar radiofrequency device (Oratec ORA-50;
Oratec Interventions) set to the standard settings of 67°C and a power of
40 W. Irrigation flow was set on low during the procedure, and the fluid was
at room temperature. The chondral surfaces were found to be normal, and no
thermal treatment was applied to the chondral surfaces. Although the patient
initially did well postoperatively, he began to have pain, decreased motion,
and decreased function approximately six months postoperatively. Twelve months
following the thermal capsulorrhaphy, an examination demonstrated forward
elevation to 60°, external rotation to —10°, and internal
rotation to the side. Radiographs showed complete obliteration of the
glenohumeral joint space (Figs. 6-A and
6-B). The results of laboratory tests, including the white
blood-cell count, measurement of the erythrocyte sedimentation rate, and
measurement of the level of C-reactive protein, were normal, and cultures of
specimens from the glenohumeral joint were negative. The patient underwent an
arthroscopic débridement and manipulation under anesthesia, at which
time it was noted that there was complete loss of the articular cartilage of
the glenoid and a loss of >75% of the humeral head. Four months after the
arthroscopic procedure, the patient returned with increased pain, decreased
motion, and decreased function, at which time he was referred to us for
further care. We performed a modified total shoulder replacement with a
Copeland resurfacing humeral arthroplasty and a lateral meniscal allograft
glenoid resurfacing procedure. At the twelve-month postoperative evaluation,
the patient had active forward elevation of the shoulder to 160°, external
rotation in abduction to 90°, and internal rotation to the T7 level. He
had returned to normal activity without pain, and shoulder strength was
symmetric with the contralateral side.
Chondrolysis is a rare but devastating complication about which little is
currently known. Tamai et al. reported on the cases of two patients in whom
chondrolysis developed following a "color test"-assisted
rotator-cuff repair9
in which gentian violet was injected into the glenohumeral joint to facilitate
the identification of a rotator cuff tear. Gentian violet, which has been
identified as being toxic to chondrocytes, was believed to be the source of
the chondrolysis. Nakagawa et al. also implicated gentian violet as a toxic
source of chondrolysis in a single-patient case
report10.
During the application of thermal energy, the tissue is heated to 65°C
and the heat-labile collagen cross-links are damaged. The triple-helix
arrangement of the collagen polypeptide chains is unwound. Early postoperative
histological studies have shown that the treated tissue undergoes an
inflammatory response with collagen fusion, pyknosis of fibroblasts, and
vascular thrombosis. The tissue also undergoes a repair response with synovial
hyperplasia and fibroblast
proliferation11.
Some authors12 have
reported that thermal chondroplasty can be safely used to smooth roughened
articular surfaces and areas of chondromalacia without destroying
chondrocytes. However, others have shown that monopolar and bipolar
radiofrequencies applied to bovine and human articular cartilage can result in
chondrocyte
death13,14.
Since chondrocytes have a very limited ability to regenerate, these authors
have recommended limiting the clinical application of these modalities,
especially bipolar radiofrequency. Furthermore, Lu et
al.13 reported that
lavage temperature may influence chondrocyte death. In their ex vivo study,
decreased chondrocyte death was seen when the lavage solution was at a
temperature of 37°C compared with the standard room temperature of
22°C.
Petty et al.6
recently reported on three cases of rapid glenohumeral joint destruction
following arthroscopy of the shoulder. A sixteen-year-old boy underwent
extensive synovectomy and resection of the coracoacromial ligament and a
modified subacromial decompression with use of radiofrequency ablation.
Because the patient had continued pain, radiographic changes, and loss of
motion, repeat arthroscopy was performed eight months after the index
procedure, demonstrating complete loss of the glenoid articular cartilage. The
articular cartilage of the humeral head was intact, but marginal osteophytes
were present. The second patient in their series had an arthroscopic
subacromial decompression and undersurface rotator-cuff débridement
with insertion of a pain pump. Severe glenohumeral osteoarthritis was found
during a second arthroscopic procedure six months following the index
procedure. Chondrolysis developed in the third patient following thermal
capsulorrhaphy with a monopolar device. Although the cause of the chondrolysis
remains unclear from this report, thermal treatment was used in all three
patients; however, thermal capsulorrhaphy was used only in the one patient who
had multidirectional instability. While all three of these patients were
successfully treated with subsequent arthroscopic débridement, the
duration of follow-up averaged only twelve months, and the long-term outlook
is unknown.
The cause of the chondrolysis in our two patients appears to be related to
previous thermal capsulorrhaphy. In both instances, the procedures were
performed by experienced shoulder surgeons and a review of the operative
images and reports showed no deviation from the standard of care. One of the
patients underwent a second arthroscopic procedure, but that procedure failed
and the patient had continued pain, loss of motion, and substantial functional
loss. Both patients eventually required joint replacement to treat the
advanced glenohumeral destruction. Although the long-term results of joint
resurfacing and biological resurfacing with meniscal allograft are unknown, we
believed that these procedures offered both of our patients the best potential
outcome.
Although this report presented the cases of only two patients, it may be
appropriate to be concerned about the effects of thermal capsulorrhaphy, not
only because of its previously
described1,2,4,6
high rate of complications, but also because of the unknown effects that
thermal treatment may have on chondrocytes. Additional long-term studies
should be performed to investigate the link between thermal capsulorrhaphy and
chondrolysis, and the devastating complications presented in this report
should cause serious consideration with regard to the use of thermal
capsulorrhaphy. ?