After Total Knee
Arthroplasty And Acupuncture: Case Report
And Survey Of Medical Acupuncturists
Steven E. Braverman, MD
Rafael L. Prieto, MD
Background While acupuncture is a safe treatment modality, complications
can occur. The incidence of sequelae in patients with total joint replacement
who undergo acupuncture therapy is not well known.
Objectives To describe a patient receiving acupuncture with sequelae
after arthroplasty, and to establish the incidence of such complications
in medical acupuncture practice.
Design Case report, and survey of medical acupuncturists (n=300).
Main Outcome Measure Incidence of complications following acupuncture
in patients receiving total joint replacement.
Results The patient developed Staphylococcus aureus infection
in the knee 7 weeks postoperatively. Acupuncture was discontinued and
the infection resolved with intravenous antibiotics, debridement, and
revision arthroplasty. Response rate to the survey was 42% (n=126).
Of the respondents, 103 (81.7%) performed acupuncture for patients who
had undergone arthroplasty. Regardless of the number of patients treated
per year, no acupuncturists reported infectious sequelae in this population.
Conclusion From our survey results, it appears that the risk
of infectious complications from acupuncture is extremely low. It is
possible in our case that the infection was coincidental because it
occurred during the period of highest postoperative risk.
Acupuncture, Arthroplasty, Complications, Total Joint Replacement
To our knowledge, there are no prospective studies that specifically
address acupuncture risk and complications. Recent review articles demonstrate
that while acupuncture is generally safe, complications can and do happen.1,2
Herein, we describe a patient who developed a deep infection following
total knee arthroplasty several weeks into acupuncture treatment for
a comorbid condition. Subsequently, a survey was sent to medical acupuncture
practitioners in an attempt to determine practice patterns and the risk
of infection in patients with total joint prostheses who are treated
A 59-year-old man was referred to our clinic 3 weeks after bilateral
total knee arthroplasty. The patient's chief complaint and reason for
referral involved a postoperative exacerbation of his chronic low back
pain. Physical examination and magnetic resonance imaging (MRI) findings
suggested lumbar spinal stenosis as the likely cause of his back pain.
The patient expressed an interest in non-pharmacological management
of pain; thus, acupuncture was presented as an option. Patient consent
was obtained; acupuncture treatment was initiated and consisted of a
series of 3 weekly treatments. Disposable sterile needles were placed
in all selected points superficial to the muscle fascia and then coupled
in standard fashion to a low-voltage electrical stimulation device.
The bilateral acupuncture points selected for analgesia included 2 posterior
and superficial to the knee joint, KI 10 and BL 40 (Figure
The patient developed left knee pain 7 weeks following the operation.
On physical examination, a purulent exudate drained from the anterior
surgical wound. Subsequent cultures were positive for Staphylococcus
aureus. Acupuncture was discontinued prior to the 4th scheduled treatment.
Orthopedic treatment included intravenous antibiotics, surgical debridement
of the left knee, and removal of the prosthesis. A revision of the left
total knee arthroplasty was completed with good results. Initial anesthesia
was delivered by spinal block. Left knee anterior wound erythema and
a serosanguinous discharge was noted at 2 to 3 weeks postoperatively.
The patient's low back pain resolved during the course of intravenous
A survey of 4 questions was prepared and sent to 300 practice members
of the American Academy of Medical Acupuncture (AAMA). Nominal and ordinal
data were analyzed using standard methods.
The study included 126 respondents (42%) to the survey. A list of
the survey questions and responses is included in Table
1. The majority (n=103 [81.7%]) of these physicians performed acupuncture
on patients who had received either a total knee or hip arthroplasty,
and 87 (84.5%) of these clinicians placed needles in the vicinity of
the surgical site as part of their treatment.
Fifty-two physicians (50.5%) performed 10 or fewer treatments in the
vicinity of the replaced joint, while 44 (42.7%) reported performing
between 11 and 100 of these treatments a year. Seven physicians (6.8%)
performed acupuncture in the vicinity of a replaced joint in more than
100 patients a year. There were no joint infections reported during
acupuncture treatment. The survey did not request information regarding
the timing of acupuncture in relation to the patients' surgeries.
The efficacy of acupuncture in the treatment of many diagnoses, and
its safety when properly performed, were discussed in a recent statement
released by the National Institutes of Health (NIH).3
Similar to conventional medical therapies, acupuncture involves some
degree of risk. Literature reviews of acupuncture-associated adverse
effects have grouped the majority of these events into 3 main categories:
mechanical organ injuries (pneumothorax, spinal cord injury), infections
(hepatitis, endocarditis, HIV), and others (asthma exacerbation, dermatitis).1,2
Many of the mechanical injuries are sustained from the technique of
Japanese or implantation acupuncture. In this process, needles are inserted
into the body and then cut. Injury of the spinal cord, upper urinary
tract, and other organs have occurred when the needles migrate to these
The transmission of infectious disease is a potential complication whenever
acupuncture needles are reused. The use of sterilized needles by appropriately-trained
staff is the crucial first step in preventing infectious complications.
Such complications of acupuncture have been reported only in cases in
which sterilization was inadequate, in-dwelling needles were used, or
the patient had a pre-existing medical condition commonly associated
with immune system compromise.1,5,6 None of the 3 types of sequelae
applied to this case where sterile, disposable needles were coupled
to a low-voltage stimulation device that may impart some mild bactericidal
|Table 1. Survey of Medical
|1. Do you treat patients with acupuncture
who have had total knee or hip replacement?
|| 23 (18.3)
|2. If yes, do you use points in the vicinity
of the replaced joint (including scar area)?
|Did not answer
|3. Approximately how many such treatments
have you performed per year?
|| 7 (6.8)
|4. How many of these joints becameinfected
during your treatment period?
|1% of the time
|1%-10% of the time
|>10% of the time
|* All data are presented as No. (%) of respondents.
The denominator for questions 2-4 is 103 acupuncturists, since 23
respondents reported "no" to question 1.
The incidence of deep infection following total knee arthroplasty is
approximately 1%.7,8 Variation in the rate of infection occurs among
medical centers, with 1 center in 1990 reporting an incidence of infection
during a 3-year period as high as 8.2%.9
Perioperative antibiotic prophylaxis following total knee arthroplasty
is currently standard practice to prevent early postoperative infection.10
One source defines early postoperative infection as infections occurring
in the first 2-3 weeks, and late infection as those occurring thereafter.10
As many as 62% of infections occur in the late postoperative period.11
There are no MEDLINE reports of acupuncture as a possible etiology of
wound infection following total knee or hip arthroplasty. The physicians
who responded to our survey frequently performed acupuncture in the
vicinity of a total joint replacement. Although the number of such patients
who are treated without infectious sequelae could not be precisely determined
from this study, we estimate that between 901 and 8,570 patients were
safely treated with acupuncture following total joint replacement. If
acupuncture contributed to the reported infection, then the overall
patient survey data indicate that the risk of infection when performing
acupuncture in these patients may be greater than zero, but much less
than 0.1%. The large number of patients with total joint replacement
who received acupuncture treatment without developing postoperative
infection, coupled with the recognized total knee arthroplasty infection
rate of 1%-8%, raises the possibility that this infection was coincidental
and unrelated to the acupuncture.
The postoperative timing of acupuncture intervention in the surveyed
treatments is unknown. In our case, the acupuncture intervention occurred
during the highest risk period for infection. That timing increases
the likelihood that a coincidental infection could be attributed to
One can neither definitively conclude nor refute that acupuncture was
the source for this S aureus infection. Acupuncturists should avoid
needling the extremity containing the prosthetic joint during the immediate
postoperative period (3-6 weeks) to decrease any potential infection
risk, and decrease the likelihood that a postoperative coincidental
infection would be attributed to acupuncture. Prospective, randomized,
controlled investigations are needed to more accurately quantify the
risks of acupuncture.
AJ. Adverse effects of acupuncture: a study of the literature for
the years 1981-1994. J Altern Complement Med. 1996;2:291-297.
E, White A. Life-threatening adverse reactions after acupuncture?
a systematic review. Pain. 1997;71:123-126.
Conference. Acupuncture. JAMA. 1998;280:1518-1524.
K, Nishio A, Nishikawa M, Ohinata Y, Sakaguchi M, Nishimura S. Subarachnoid
hemorrhage and spinal root
MG. Fatal Staphylococcal septicemia following acupuncture: report
of two cases. R I Med J. 1982;65:251-253.
S, Ohi T, Sugimoto S, Satoh S, Matsukura S. Cervical spinal epidural
abscess following acupuncture: successful treatment with antibiotics.
Intern Med. 1998;37:161-165.
KL, Hanssen AD. Infection after total hip arthroplasty: past, present,
and future. J Bone Joint Surg Am. 1995;77:1576-1588.
MG, Kelley K, Thornhill TS. Infection as a complication of total
knee-replacement arthroplasty. J Bone Joint Surg Am. 1990;72:878-883.
SM, Culver DH, Simmons BP, Jarvis WR. Risk factors for wound infections
after total knee arthroplasty. Am J Epidemiol. 1990;131:905-916.
WN. The Knee. St Louis, Mo: Mosby-Year Book; 1994:1262.
AT Jr, Tsukayama D, Gustilo RB. Effect of time of onset and depth
of infection on the outcome of total knee arthroplasty infections.
Clin Orthop. 1991;273:98-104.,
Lieutenant Colonel Steven E. Braverman, MD, is on active duty in the
United States Army, and Deputy Commander for Clinical Services, Moncrief
Army Community Hospital, Ft Jackson, South Carolina. The research for
this study was done at Walter Reed Army Medical Center, Washington,
DC, where Dr Braverman was the Physical Medicine and Rehabilitation
Residency Program Director and Chief of the Physical Medicine and Rehabilitation
Lieutenant Colonel Steven E. Braverman, MD
Deputy Commander for Clinical Services
Moncrief Army Community Hospital
4500 Stuart St
Ft Jackson, SC 29207
Phone: 803-751-2280 (DSN 734) o Fax: 803-751-2784
Major Rafael L. Prieto, MD, is on active duty in the United States Army,
and is a Physiatrist at Landstuhl Regional Medical Center in Germany.
Major Rafael L. Prieto, MD
Physical Medicine and Rehabilitation
Department of Surgery
Landstuhl Regional Medical Center, Germany
Mailing Address: MCEUL-CMR 402, APO, AE 09180