Three patients of 222 who had total knee arthroplasty (TKA) with
peripheral nerve blocks (PNB) suffered a post-operative
in-hospital fall with complete wound disruption requiring
emergency re-operation. These complications all occurred late in
our experience with PNB and followed a change in ward nursing
organization. All three patients attempted to ambulate to or
from the bathroom without assistance despite instruction to the
knee arthroplasty; peripheral nerve block; complications;
Peripheral nerve blockade (PNB) for total knee arthroplasty
decreases pain, motor function, and proprioception in the
involved limb. It has been reported to lessen
postoperative use of narcotics and their attendant side effects
[1,2,3], lower pain scores [2,4], speed recovery [2,4,5], and
lower costs  but may give rise to serious complications .
Nerve injuries are uncommon with use of ultrasound guidance and
nerve stimulation to increase the anatomic and physiologic
accuracy of injection [7,8]. Other complications have been
We report three patients that sustained falls with complete
traumatic wound dehiscence within hours of successful
administration of PNB for TKA. This unanticipated complication
of new technology has been reported by others in small numbers
[7-9]. Possible causes and methods of avoiding this severe
complication are discussed.
Three patients of 222 had falls with complete wound dehiscence
following TKA. For the first 9 months of our experience (105
patients) continuous femoral nerve block (FNB) by pump with an
indwelling catheter was used. During the following 10 months
(117 patients) a single-dose sciatic nerve block (SNB) was added
to address posterior pain [10, 11]. Preoperatively, informed
consent was obtained and all patients were carefully instructed
that they must not ambulate after surgery until otherwise
Ultrasound and electrical stimulation confirmed that the
catheter was properly placed near and not in the femoral nerve.
A constant infusion of 0.2% ropivacaine was administered at a
rate of 6 ml/hr by commercial external pump designed for this
purpose. The catheter was left in place the day of surgery and
through the first postoperative day. The rate was adjusted down
to 2 ml/hr at the nurse’s discretion. Supplemental parenteral
and/or oral narcotic medication was provided as necessary.
When SNB was used, the sciatic nerve was identified
approximately 15 cm proximal to the posterior knee. Ultrasound
and electrical stimulation were used as above to guide a single
preoperative 20 ml injection of 0.25% bupivacaine .
On postoperative arrival to the ward, patients were again
reminded that they must not ambulate unassisted until otherwise
instructed. With diminution of anesthetic effect after
scheduled removal of the catheter and pump, a physical therapist
experienced in working with PNB TKA patients evaluated the
patient for readiness for ambulation by a defined protocol.
First, the patient is asked to perform a
activation ("quad set"). If satisfactory, the patient is
assisted to sit at the bedside and asked to perform a "large
arch" range of motion of the knee. If acceptable, the patient
is asked to attempt to stand without physical assistance, using
the walker. If unable, or if physical therapist examination
determines that altered proprioception still exists, the effort
is deferred. If each of these elements is evaluated as
acceptable by the physical therapist, ambulation proceeds with
the use of the walker under the careful guidance of the physical
therapist. Otherwise, the attempt is terminated and
re-evaluation is scheduled for a later time.
Case 1: A 56-year-old woman with traumatic arthritis and valgus
deformity required use of a walker. She had failed conservative
treatment, including NSAIDs and injections of hyaluronic acid
and cortisone. She had unilateral TKA with PNB of the femoral
and sciatic nerves as above. Eight hours after her surgery she
tried to go to the bathroom unassisted. She had inadequate
control of the leg. She fell and flexed her knee with complete
wound dehiscence. She had not asked for assistance.
Case 2: A 41-year-old woman with bilateral degenerative
arthritis with mild varus deformity of the knees was planned for
left TKA with FNB and SNB. She used a cane and failed the same
conservative maneuvers as Case 1. A physical therapist assisted
her to walk to the bathroom, and then was left alone for
privacy. The patient attempted to get off the toilet unassisted
and fell to the floor due to lack of leg control. The wound
Case 3: A 72-year-old man with right knee arthritis, varus
deformity, and significant heart disease used a cane and had
failed conservative maneuvers, as Case 1. He underwent TKA with
FNB and SNB. Postoperatively, he tried to walk unassisted to
the bathroom. The nurse was standing at the door of the room.
His knee flexed as he fell to the floor with the same wound
dehiscence as Cases 1 and 2 and was found to have a peroneal
nerve dysfunction after the fall.
In each case after their injury the patient recalled the
admonition to not walk without assistance. Post-fall, each
patient was treated identically. The wound was immediately
packed in Betadine antiseptic-soaked gauze, parenteral
antibiotics were administered, and the patient was taken back to
the operating room where the knee and soft tissue was irrigated
with 10 liters of sterile saline, 200 cc Betadine and Bacitracin
(antibiotic) solution, and the wound was primarily reclosed.
Prophylactic IV antibiotics were switched after 3 days to oral
antibiotics, which were continued for two weeks. Intraoperative
cultures were negative for bacterial growth in all 3 cases. Each
patient was informed that data concerning their case would be
submitted for publication.
At early follow-up (6, 7, and 8 months for Case 1, 2 and 3,
respectively) none of these patients had developed a superficial
or deep infection. All had primary wound healing. Case 2 had
restriction of range of motion of 0-30 degrees six weeks after
her repair. This was treated with closed manipulation under
general anesthesia resulting in knee range of motion to 0-100
degrees 3 months after initial surgery. Case 3 had an extended
hospital stay due to cardiac issues unrelated to his knee
surgeries. His peroneal nerve dysfunction was not completely
resolved six months after the surgery.
Sequelae of falls with dehiscence reported to date include loss
of range of motion, fracture, and prolonged drainage, but
fortunately not post-operative infection . In the
report of major complications associated with FNB by Feibel, et
al,  8 of 1190 (0.6%) TKA patients had falls of which only 3
(38%) had wound dehiscence. This group has subsequently reduced
usage of PNB in favor of periarticular injections. Atkinson, et
al  reported four cases of patient falls and Kandasami, et al
 reported five cases of falls of a cohort of 235 (2.1%).
These authors suggest that the loss of pain, proprioception and
motor control were responsible for the falls. Whiting, et al,
 reported sciatic perineural use of clonidine and
buprenorpine in 2 cases of knee surgery with preservation of
below-knee gross motor function.
Fall with complete wound dehiscence was first seen rather late
in our PNB experience. For the first 191 patients over 16
months, no such complications occurred. This was followed by 3
occurrences in 31 patients over a 3-month period following a
change in nursing unit organization. In an endeavor to avoid
future problems of this type, we evaluated possible contributing
factors to patient falls.
Addition of SNB.
After 9 months of using FNB only (105 patients) we adopted the
addition of single-dose sciatic nerve blocks to address residual
posterior pain [10,11]. This resulted in a nearly equal number
of FNB only and FNB + SNB patients (105 and 117 respectively).
There were no falls with FNB only or during our first 7 months
(159 patients) after addition of SNB but before the change in
nursing organization (see below).
It seems unlikely, therefore, that addition of SNB precipitated
our cluster of cases.
Adequacy of pre-ambulation evaluation.
An inadequate evaluation by personnel could conceivably lead to
premature permission to ambulate. We utilized physical
therapists well experienced and trained in the evaluation of PNB
TKA patients. Both the nurses and the patients themselves agreed
that the three case patients had not yet received permission to
stand or ambulate without assistance, so this possible factor
Patient failure to follow instructions.
A factor common to these patients’ falls was failure of the
patient to follow instructions to avoid ambulation without
assistance. In each case, after their injury the patients
acknowledged they had received this instruction. Patient 1 and 3
fell when they attempted to get to the bathroom without
assistance. Case 2 had assistance from bed to bathroom, where
she was left alone for privacy. When done with the toilet, she
attempted to stand without calling for assistance and fell. The
PNB likely caused these patients to overestimate their
Medications, strange surroundings, and fat embolism may
contribute to confusion and failure to follow postoperative
instructions. In such case, strict bed rest with restraints may
be necessary to enforce the instruction and thus avoid falls,
but none of these patients was noted to have such problems in
the nursing record or on post-fall evaluation. These patients
were in possession of their faculties but misinterpreted their
capabilities, probably as a result of their PNB treatment. The
complications associated with restraints, their unpleasant
patient experience, and the expense of the burdensome
administrative overhead [13, 14] argue against their routine use
but rather for close patient observation with very selective
It may be thought that use of a knee immobilizer would help
patients stay in bed and/or prevented leg buckling and falls.
However patients and staff may overestimate the adequacy of its
support. This may inappropriately encourage premature
ambulation, thus actually increasing rather than reducing the
risk of falls. The immobilizer also impedes the use of CPM.
For these reasons, we prefer not to use knee immobilizers.
Urinary catheter use.
A factor common to the three cases was the patients’ need to
empty their bladders. Routine use of urinary catheters would
have eliminated the ill-fated desire to ambulate to the
bathroom. Complications of urinary catheters are common, can be
catastrophic (bacteremic seeding of the prosthesis, for
instance) and such complications may not be reimbursable under
new Medicare rules [15-17]. We do not routinely use urinary
drainage catheters, but prefer prompt response to patients’
requests for assistance by trained and experienced personnel.
Increased nursing awareness of patients' disability with PNB and
the consequences of not meeting their bedpan needs might have
diminished the risk of ill-fated attempts to walk to the
bathroom. Our new motto is, “Don’t pee with your new knee”.
For the first 16 months of our PNB use, our hospital had a
dedicated Total Joint Arthroplasty (TJA) Program. On the ward,
this consisted of designated adjacent rooms, a dedicated group
of nurses and a physical therapist. A mandated change resulted
in our patients no longer having designated personnel or rooms
for the final 3 months and 31 patients of the series. The
patients were then cared for in a general orthopedic ward.
Retrospective review revealed that 0 of 191 patients with the
TJA program and 3 of 31 patients subsequent to its termination
experienced falls with wound dehiscence. Case 1 occurred two
weeks after the change.
The use of PNB and our desire to gain the benefits of early
mobilization may not be wholly congruent. Perhaps it is too
much to ask of our patients and staff. We remain attracted to
the benefits of PNB for TKA as described in the literature,
despite the increased risk of falls due to decreased motor,
proprioception and pain sensations. It did not appear that
addition of SNB was associated with the falls. Alterations in
type, dosage, duration or concentration of injected perineural
medications might decrease the risk of falls but this is yet to
be elucidated. Routine use of restraints, knee immobilizers, or
urinary drainage catheters in TKA patients does not appear
appropriate at present. While they might diminish falls and
compensate for inattentive nursing, the associated morbidity and
delayed mobilization would be unattractive to patients and
Patient selection is important. Use of PNB in the patient that
is relatively weak in relation to his habitus or severely
deconditioned as a result of arthritis, or who has neurologic or
balance problems may not be appropriate. Suspicion that the
patient cannot or might not follow instructions closely should
preclude them from PNB.
Patient education appears crucial yet was ineffectual in these
three cases. All three patients reported that they had been
instructed properly. We have redoubled efforts at patient
education, making sure we have thoroughly discussed the risk of
this severe complication with each patient, to impress upon them
the importance of the ambulation instructions. We ask the
patients to weigh with us the risk of fall and dehiscence with
PNB, to gain the benefits of reduced pain and diminished
narcotic use. We recognize the important role of specifically
trained nurses in reinforcing patient education.
Our experience suggests that the most important factor in
preventing this complication is an adequately staffed, trained
and committed team. Noting the temporal association of the
complication cluster with disbanding of the TJA service, we are
reconstituting this service, to regain its perceived benefits.
With a dedicated staff repeatedly working with the same
physician(s), in the same rooms, with the same orders, the same
type of anesthesia/analgesia, and type of operative procedure,
this serious complication was absent during the first 19 months
of our 22 month series. Neither physician’s orders alone nor
the use of restraints, catheters, or immobilizers are likely to
equal a dedicated TJA nursing team that is vigilant and
responsive to these particular patients’ subtle needs, and to
the unintended consequences of new technology.
De Ruyter ML, Brueilly KE, Harrison BA, Greengrass RA, Putzke
JD, Brodersen MP. A pilot study on continuous femoral
perineural catheter for analgesia after total knee
arthroplasty: the effect on physical rehabilitation and
outcomes. J Arthroplasty. 2006;21:1111.
Kadic L, Boonstra MC, DE Waal Malefijt MC, et al. Continuous
femoral nerve block after total knee arthroplasty? Acta
Anaesthesiol Scand. 2009 53:914-20.
Allen HW, Liu SS, Ware PD, et al. Peripheral nerve blocks
improve analgesia after total knee replacement surgery. Anesth
Singelyn FJ, Deyaert M, Joris D, et al. Effects of
intravenous patient-controlled analgesia with morphine,
continuous epidural analgesia and continuous three-in-one
block on postoperative pain and knee rehabilitation after
unilateral total knee arthroplasty. Anesth Analg 1998;87:88.
Motamed C, Combes X, Ndoko SK, et al. Effect of pre-incisional
continuous regional block on early and late postoperative
conditions in tibial osteotomy and total knee arthroplasty.
Open Orthop J. 2009;3:22.
Ilfeld BM, Mariano ER, Williams BA, et al. Hospitalization
costs of total knee arthroplasty with a continuous femoral
nerve block provided only in the hospital versus on an
ambulatory basis: a retrospective, case-control, and
cost-minimization analysis. Reg Anesth Pain Med. 2007;32:46.
Feibel RJ, Dervin GF, Kim PR, et al. Major Complications
Associated with Femoral Nerve Catheters for Knee Arthroplasty.
J Arthroplasty. 2009;24:132.
Kandasami M, Kinninmonth AW, Sarungi M, et al. Femoral nerve
block for total knee replacement - a word of caution. Knee
Atkinson HD, Hamid I, Gupte CM, et al. Postoperative fall
after the use of the 3-in-1 femoral nerve block for knee
surgery: a report of four cases. J Orthop Surg (Hong Kong).
Hunt KJ, Bourne MH, Mariani EM. Single-injection femoral and
sciatic nerve blocks for pain control after total knee
arthroplasty. J Arthroplasty. 2009;24:533.
Ben-David B, Schmalenberger K, Chelly JE. Analgesia after
total knee arthroplasty: is continuous sciatic blockade needed
in addition to continuous femoral blockade? Anesth Analg.
Whiting DJ, Williams BA, Orebaugh SL, et al. Case report:
postoperative analgesia and preserved motor function with
clonidine and buprenorphine via a sciatic perineural catheter.
(J Clini Anesth. 2009; 21:297.
Pappas SH. The cost of nurse-sensitive adverse events. J Nurs
Brandeis GH, Ooi WL, Hossain M, et al. A longitudinal study of
risk factors associated with the formation of pressure ulcers
in nursing homes. J Am Geriatr Soc. 1994;42:388.
Saint S, Meddings JA, Calfee D, et al. Catheter-associated
urinary tract infection and the Medicare rule changes. Ann
Intern Med. 2009;150:877.
Nazarko L. Providing effective evidence-based catheter
management. Br J Nurs. 2009;18:S4.
Patel HR, Arya M. The urinary catheter: 'a-voiding
catastrophe'. Hosp Med. 2001;62:148.