Patient Demographics
We performed a retrospective review of the cases of all patients who had
undergone proximal row carpectomy for the treatment of degenerative wrist
conditions between 1980 and 1992, thereby providing a minimum follow-up period
of ten years. The study was approved by our institutional review board.
During the study period, proximal row carpectomy was performed in thirty
patients. Three patients died, three refused to participate in the study, and
three were lost to follow-up, despite a review of medical records and multiple
Internet searches. This left twenty-one patients (seventeen men and four
women) with a total of twenty-two wrists available for the study (see
Appendix). The average age at the time of the operation was thirty-eight years
(range, nineteen to fifty-seven years). The dominant hand was involved in
fifteen patients. Ten patients were laborers. The etiology of the degenerative
arthritis included scapholunate advanced collapse (nine wrists), scaphoid
nonunion with advanced collapse (six), and Kienböck disease (seven). All
of the wrists with scapholunate advanced collapse and scaphoid nonunion with
advanced collapse had radiographic evidence of radioscaphoid arthritis with
preservation of the capitolunate articulation. Those with Kienböck
disease had fragmentation of the lunate with carpal collapse. No wrist had
radiographic evidence of capitolunate arthritis or degenerative changes
between the distal part of the radius and the lunate.
Pain and weakness were the primary causes of reduced function in all
patients. Conservative treatment consisting of bracing, activity modification,
anti-inflammatory medication, and in some cases corticosteroid injections had
failed for all patients.
A variety of procedures had been performed prior to the proximal row
carpectomy in fifteen wrists, with four wrists undergoing multiple procedures.
The procedures included open reduction and internal fixation of a scaphoid
fracture (two), bone-grafting for scaphoid nonunion (four), prosthetic lunate
replacement (three), radial shortening osteotomy (one), excision of loose
bodies (three), arthroscopy (one), arthrotomy with drilling of the scaphoid
(one), scapholunate ligament repair with dorsal capsulodesis (one), and
intercarpal fusion (one). Preoperative radiographs revealed radioscaphoid
arthritis in all patients. No wrist had radiographic evidence of arthritis at
the capitolunate joint.
Of the twenty-one patients (twenty-two wrists) available for follow-up, all
but one were able to return to the office for an examination and radiographs.
The patient who was unable to return was contacted by telephone, and the
results of the two questionnaires were recorded. The office charts and
radiographs were reviewed. Subjective information was obtained with use of two
questionnaires. The Disabilities of the Arm, Shoulder and Hand (DASH)
Questionnaire23 was
chosen to measure disability of the upper extremity. The DASH is a thirty-item
validated outcomes questionnaire, divided into two parts, that assesses
physical function and symptoms. It is scored from 0 to 100 points, with higher
scores representing greater disability. We devised a second questionnaire to
assess patient satisfaction, severity of wrist pain, use of medication, and
the ability to return to previous employment (see Appendix).
Objective measurements included the range of motion of the wrist (flexion,
extension, radial deviation, and ulnar deviation measured with a handheld
goniometer) as well as grip strength (measured with a Jamar dynamometer
[Sammons Preston, Bolingbrook, Illinois]). Radiographs of the wrist (true
posteroanterior and lateral views made with the elbow flexed 90° and the
forearm in neutral rotation) were also made. All radiographs made since 1986
were made in one office with the same equipment and technique. The
radiocapitate space was measured, and the presence of cysts, sclerosis,
osteophytes, and capitate flattening was noted. A radiographic classification
based on the amount of narrowing of the radiocapitate space (none, partial, or
complete loss of the space) compared with that on the first postoperative
radiograph was developed (Figs.
1-A,
1-B,
2-A,
2-B,
3-A,
3-B). All three of us
independently reviewed the radiographs, and an average rating was obtained.
The average duration of follow-up was fourteen years (range, ten to
twenty-three years; median, thirteen years).
Surgical Technique
The carpus was approached through a dorsal longitudinal incision. The
extensor retinaculum between the third and fourth extensor compartments was
then opened. The posterior interosseous nerve was not routinely resected. The
capsule was opened longitudinally, and flaps were raised ulnarly and radially.
The lunate fossa and the articular surface of the proximal pole of the
capitate were inspected to be sure that there was little or no chondromalacia.
If extensive degenerative changes were noted, a different surgical procedure
was carried out (i.e., intercarpal or total wrist arthrodesis). The proximal
carpal bones were individually removed. If radial deviation was found to be
limited intraoperatively, a radial styloidectomy was performed (done in ten of
the twenty-two wrists). Early in the series, a 1.6-mm Kirschner wire was
placed temporarily across the radiocapitate joint to maintain length, secure
alignment, and facilitate healing in the early postoperative period. Care was
taken not to place the Kirschner wire across the articular surface of the
capitate. Later, this step was thought to be unnecessary and was abandoned
after complications, including pin-track infection, were encountered. The
capsule and retinaculum were reapproximated with 2-0 nonabsorbable sutures.
The extensor pollicis longus was not routinely transposed. The wrist was
immobilized in a volar splint in 10° of extension after the surgery.
Motion of the digits was begun immediately after the surgery. The Kirschner
wire, when used, was removed at four weeks, at which time a protected range of
motion of the wrist was initiated. All immobilization was discontinued at six
weeks postoperatively.
Statistical Analysis
Statistical analysis was performed with use of Student t tests, analysis of
variance, and chi-square tests. Differences were regarded as significant at p
= 0.05.
There were four failures (18%), all with severe wrist pain. All four wrists
underwent arthrodesis (Table
I), at an average of seven years (range, eight to 140 months;
median, nine years) after the proximal row carpectomy. The average age of the
four patients at the time of the proximal row carpectomy was twenty-nine
years, and none were more than thirty-five years of age (p = 0.03). The
underlying diagnoses were scaphoid nonunion with advanced collapse in two of
the wrists that failed and Kienböck disease in the other two. None of the
nine wrists that underwent proximal row carpectomy as a result of scapholunate
advanced collapse required arthrodesis. Of the four wrists converted to a
fusion, three had undergone at least one procedure prior to the proximal row
carpectomy. The preoperative diagnosis was considered to be work-related in
three of these four patients. A radial styloidectomy had been performed in two
of the patients; one had been done prior to the proximal row carpectomy and
the other, at the time of the proximal row carpectomy.
The first failure occurred in a twenty-eight-year-old male construction
worker eight months postoperatively. This patient had previously undergone a
contralateral wrist arthrodesis and requested conversion of the proximal row
carpectomy to an arthrodesis. The second failure occurred in a woman with a
history of silicone synovitis after failure of a lunate implant that had been
placed for the treatment of Kienböck disease. She underwent conversion to
a wrist arthrodesis at eighty-one months. The remaining two failures occurred
in young laborers more than ten years (130 and 140 months) after the proximal
row carpectomy. Both patients reported more than nine years of pain relief
after the proximal row carpectomy, prior to the gradual progression of pain
leading to the wrist arthrodesis.
Three patients did not return to their previous occupation after the
proximal row carpectomy. Two had permanent restrictions, and one was limited
to light duty. All of the remaining eighteen were able to return to their
previous employment without restrictions.
The fifteen wrists in the fifteen patients with unilateral involvement in
whom the procedure did not fail and who were available for follow-up
examination had an average arc of flexion-extension of 72°. They had an
average of 36° of flexion, 36° of extension, 9° of radial
deviation, and 31° of ulnar deviation, or an average of 62%, 60%, 47%, and
73% of the values for the contralateral wrist, respectively. Grip strength
averaged 91% of that on the contralateral side.
The patients were very satisfied with the results of the proximal row
carpectomy in fourteen of the eighteen wrists that did not fail and were
satisfied with the results in the remaining four wrists. No patient was
dissatisfied or very dissatisfied. Of the eighteen wrists, nine were reported
by the patients to be not painful; four, as mildly painful; and five, as
moderately painful. No patient reported severe wrist pain. Six wrists required
the patient to use pain medication sometimes, and one required it every day.
The remaining eleven wrists did not require pain medication. Finally, the
average DASH score for the eighteen wrists (seventeen patients) that did not
fail was 9 points, with 100 points representing maximal disability.
A review of the radiographs revealed degeneration of the radiocapitate
articulation in all but three of the seventeen wrists for which the studies
were available. Narrowing of the radiocapitate joint space was rated as
partial in seven and complete in seven. The average width of the radiocapitate
space was 1 mm on the posteroanterior and lateral radiographs. Despite the
extent of the degeneration noted on the radiographs of the wrists that had not
failed or required arthrodesis, no significant association was found, with the
numbers available, between the degree of radiograph degeneration and the age
or gender of the patient, range of motion, level of wrist pain, use of pain
medication, work restrictions, or patient satisfaction (p > 0.05). There
were two complications: both were pin-track infections that resolved with a
short course of oral antibiotics. There were no neurovascular injuries or
wound complications.
Proximal row carpectomy is an accepted method for the treatment of
degenerative arthritis of the wrist. Preservation of a functional range of
motion, adequate grip strength, and high levels of patient satisfaction have
been reported in several peer-reviewed reports. The ranges of
flexion-extension and grip strength, expressed as percentages of the values on
the contralateral side, have averaged 62% and 80%, respectively, with patient
satisfaction ranging from 80% to
100%2-15,21,22.
Low failure rates, ranging from 0% to 20%, have also been
noted3-5,8,9.
With the exception of two
reports21,22,
the follow-up periods in these
studies2-15
have averaged less than nine years.
In reports with longer
follow-up21,22
and in the present study, radiographic evidence of narrowing of the
radiocapitate articulation has been noted and is of concern with respect to
long-term patient satisfaction. Imbriglia et
al.8 measured the
radii of curvature of the capitate and the lunate fossa of the distal part of
the radius on serial radiographs and found the radius of curvature of the
capitate to be 64% of the corresponding value for the lunate fossa on the
preoperative anteroposterior radiograph and 60% on the true lateral
radiograph. Using cineradiography, they also found that motion of the capitate
on the radius is translational, with a moving center of rotation. The
combination of size mismatch between the capitate head and the lunate facet of
the distal part of the radius and the translational motion of the wrist
following proximal row carpectomy probably is responsible for the loss of the
radiocapitate cartilage space seen on long-term follow-up radiographs.
In a follow-up to the 1990
series8,
Imbriglia21
reported on twenty-seven patients at an average of nine years postoperatively.
He commented that twelve of the twenty-seven patients had radiographic
evidence of measurable cartilage space narrowing, but, despite this finding,
he did not note a decrease in clinical function.
Jebson et al.22
reported on twenty patients followed for an average of 13.1 years. There were
two failures that required radiocapitate arthrodesis. The range of motion and
grip strength averaged, respectively, 63% and 83% of the values on the
contralateral side. Seventeen of the eighteen patients who did not have a
failure reported no pain or minimal pain, and the same number were satisfied
with the procedure. Degeneration of the radiocapitate space was noted to be
absent or minimal in thirteen of the seventeen wrists and moderate or severe
in four.
In the current study, radiographic signs of degeneration of the
radiocapitate space were the rule (seen in fourteen of the seventeen wrists)
rather than the exception, at an average of fourteen years postoperatively.
The average articular space decreased from 2 mm on the first postoperative
radiographs to 1 mm on the final radiographs. Seven of the wrists had complete
loss of the space. The higher proportion of radiographic signs of
deterioration noted in our study may be the result of the longer follow-up
period, suggesting that degeneration of the radiocapitate articulation can be
expected over time. Despite the high rate of degeneration noted on the
radiographs, a high proportion of the patients were satisfied or very
satisfied with the long-term result (eighteen wrists).
The results of this study with regard to grip strength, range of motion,
and patient satisfaction are compatible with those of previous
studies3-5,8,9,22.
The average flexion-extension arc was 72°, which represented 61% of the
flexion-extension arc on the contralateral side. Grip strength in our study
averaged 91% of that on the contralateral side, which is slightly higher than
the average grip strength of 80% (range, 66% to 100%) in other
studies2-15,22.
There were four failures in our series; three occurred in laborers and all
four occurred in younger patients, with an average age of twenty-nine years at
the time of the proximal row carpectomy. No failures were seen in patients
over the age of thirty-five years at the time of the carpectomy (p = 0.03).
This symptomatic deterioration in wrist function may be the result of higher
demands placed on the wrist by younger patients. Although all four patients
reported pain relief with subsequent wrist arthrodesis, our results suggest
that symptomatic radiocapitate degeneration may develop over the long-term in
young, active patients and thus such patients should be cautioned about the
potential for symptomatic degeneration after proximal row carpectomy.
We no longer perform temporary pin fixation. The risk of damage to the
capitate articular surface, the potential for sensory nerve injury or pin
migration, and the documented problems with pin-track infection led to this
decision. We have noted no untoward effects since discontinuing the use of
temporary pin fixation.
A radial styloidectomy was performed in ten of the twenty-two wrists. The
inclusion of a styloidectomy during proximal row carpectomy continues to be a
matter of surgeon preference, with the decision often based on an
intraoperative assessment of whether there is impingement of the trapezium or
the radial styloid. Previous authors have shown that the trapezium lies palmar
to the plane of the
radius8. However,
impingement may be noted, particularly when the styloid is enlarged or
elongated8. If
styloidectomy is chosen, preservation of the volar wrist ligaments should be
attempted7,24.
In conclusion, we believe that proximal row carpectomy provides
satisfactory long-term results in most patients. However, younger, active
patients appear to be at an increased risk for failure of this procedure.
Finally, while radiocapitate degeneration occurs radiographically in most
patients, it does not preclude a successful clinical result.
A table presenting patient demographics and a figure showing the patient
questionnaire developed for this study are available with the electronic
versions of this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM). ?