Contrary to widespread belief, total surgical
replacement of an arthritic shoulder joint carries no
greater risk of complications than replacement of other
major joints, a Johns Hopkins study suggests.
In fact, the Johns Hopkins researchers say, their
study shows that patients who undergo shoulder arthroplasty
to relieve chronic and significant pain can expect
significantly fewer complications, much shorter hospital
stays and less costs than patients undergoing hip or knee
Arthritis is estimated to affect more than 16 million
Americans and is a common cause of shoulder pain and
mobility loss. However, many reject shoulder joint
replacement, fearful of risks and costs. The Johns Hopkins
research team, led by Edward McFarland, director of the
Division of Adult Orthopedics at The Johns Hopkins
Hospital, analyzed anonymous patient information provided
by the Maryland Health Services Cost Review Commission, the
state's hospital-rate regulator. The team looked at all
arthroplasties performed in Maryland to alleviate
osteoarthritis pain between 1994 and 2001, including
details of 15,414 hip surgeries, 34,471 knee operations and
625 shoulder procedures.
"After looking at how all these patients fared, we
concluded that, comparatively, total shoulder surgery is
just as safe and effective as other types of
arthroplasties," McFarland said. "Lower numbers of shoulder
procedures done both regionally and nationally may indicate
that many people live with shoulder pain because they fear
that the corrective surgery is too risky or costs more than
similar procedures. But we have found that this is just not
The findings appeared in a recent online version of
the journal Clinical Orthopedics.
According to nationwide 2003 Medicare figures, 6,700
people had shoulder joints replaced that year, compared to
106,887 hip replacements and 199,195 total knee
Shoulder arthroplasty consists of placement of a metal
and plastic artificial joint similar to the ball-and-socket
construction used to substitute for hip and knee joints.
The socket, or scapula, is scraped clean and fitted with a
plastic mount, while the ball at the top of the humerus
bone in the arm is replaced with a metal implant.
Patients in the study who had shoulder surgery had far
less in-hospital postsurgical complications (7.5 percent)
compared with those patients who had their hips and knees
replaced (15.5 percent and 14.7 percent, respectively).
McFarland's team also determined that the average time a
person remained hospitalized after the surgery was lowest
for those recovering from shoulder procedures (2.42 days
for shoulder patients vs. more than four days for both the
hip and knee equivalents).
Shoulder arthroplasty is also less expensive,
according to McFarland. A shoulder replacement's total
costs, on average, are $10,351, whereas hip replacement
surgery averages $15,442 and knee arthroplasty, $14,674.
In the study period, there were no fatalities
associated with shoulder replacement, compared to 27 deaths
among hip replacement patients and 54 among knee
replacement patients. Given the small number of shoulder
surgeries actually performed, McFarland noted that the
mortality rates for the procedure were not statistically
different from the other forms of arthroplasty. But finding
no resulting deaths attributed to total shoulder
replacement in the group of patients studied did add
"clinical significance" to the idea that shoulder surgery
is relatively safer. The study's authors say they suspect
that the higher number of deaths related to hip and knee
procedures may be related to repeat surgeries stemming from
postoperative complications during the same
hospitalization, a need for blood transfusions or extended
intensive care stays.
McFarland says that most patients who are candidates
for total shoulder replacement surgery are "at the end of
their rope" trying to manage chronic pain and disability
with drugs. "Ninety-nine percent of the people who have a
shoulder replacement for arthritis get pain relief and say
that they wish they had done it sooner," McFarland said.
"This study indicates there may be little reason to