вход по аккаунту


Enterococcusan unusual cause of septic arthritis.

код для вставкиСкачать
59 1
Enterococci are a well-documented cause of
urinary tract infections, biliary tract infections, and an
aggressive form of endocarditis which is refractory to
medical therapy (1). However, enterococci rarely
cause infectious arthritis (2), and the optimal therapy
for enterococcal arthritis is unknown. We report 2
cases of enterococcal arthritis and review the literature to formulate guidelines for management of this
destructive form of infectious arthritis.
Patient 1. A 71-year-old man was hospitalized
for removal of a subdural hematoma. Urinary catheterization was required after surgery, and a rapidly
tapering course of dexamethasone was administered.
He was discharged, and in the 10 days before his
second admission he noted dysuria, without fever or
chills. His right knee became painful, swollen, and
warm for 3 days. There was no history of trauma to the
knee or previous arthritic symptoms.
From the Department of Medicine, Divisions of Kheumatology and Infectious Diseases, the University of Missouri-Columbia,
and the Veterans Administration Medical Center. Columbia, Missouri.
Samuel H. Zwillich. MD: Fellow. Division of Immunology
and Rheumatology. University of Missouri-Columbia; Bruce H.
Hamory, MD: Assistant Professor of Medicine, University of Missouri-Columbia; Sara E. Walker, MD: Associate Professor of
Medicine, University of Missouri-Columbia and Chief, Rheumatology Section, Harry S. Truman Memorial Veterans Hospital.
Address reprint requests to Sara E. Walker, MD, I I IF.
Harry S . Truman Memorial Veterans Hospital, ROO Stadium Road.
Columbia, MO 65201.
Submitted for publication September 27, 1983; accepted in
revised form January 10. 1984.
Arthritis and Rheumatism, Val. 27, No. 5 (May 1984)
At the second admission, the physical examination was remarkable for low-grade fever (37.8"C),
dulled affect and mentation, and a warm, tender right
knee with a large effusion. There were no skin lesions.
Synovial fluid aspirated from the right knee was
turbid with low viscosity. There were 59,400 white
blood cells/ml with 97% polymorphonuclear leukocytes. The synovial glucose was 35 mg/dl and the
simultaneous serum glucose was 117 mg/dl. Crystals
were absent and bacteria were not seen on Gram stain.
The urine sediment contained occasional white cells
and infrequent bacteria. Radiographs of the right knee
showed narrowing of the medial tibiofemoral compartment (Figure 1). Pyogenic arthritis was suspected and
therapy was begun with intravenous nafcillin sodium,
10 gm/day, and gentamicin, 80 mg intravenously every
12 hours. Serum pre- and post-dose levels of gentamicin were within accepted therapeutic ranges. Closed
needle drainage was required twice daily because of
rapid reaccumulation of the effusion, but for 3 days
symptoms and synovial fluid analysis results both
improved steadily.
When synovial fluid cultures grew group D
streptococcus (enterococcus), antibiotic therapy was
changed to intravenous ampicillin, 12 gm per day.
Blood cultures were sterile, and the urine culture
produced fewer than 10,000 colonies per ml of an
unidentified gram-positive coccus.
Despite high-dose ampicillin therapy and closed
drainage, enterococcus grew from serial synovial cultures. and on the fourth day the patient's symptoms
and signs worsened. Gentamicin was added to ampicillin, and 2 days later a radical synovectomy was
performed and drains were placed for continuous
irrigation. The patient remained febrile and cultures of
Figure 1. A and B, Patient. L{ radiographs of right knee on second admission, showing narrowed medial tibiofemoral
the knee remained positive for enterococcus. On the
ninth hospital day, the patient bled repeatedly from a
drainage site and the drains were removed. The knee
became warm, tender, and distended; the synovial
fluid white blood cell count was 34,000 with >99%
polymorphonuclear leukocytes, and the synovial fluid
glucose level was 2 mg/dl.
On the fourteenth hospital day, after 10 days of
combined antibiotic therapy, the knee was reexplored,
opened widely, and packed. No loculation, communicating popliteal fossa cyst, or osteomyelitis was seen.
Swab cultures of the open knee grew enterococcus for
3 days postoperatively. Subsequent cultures were sterile. Radiographs showed progressive loss of joint
space and periosteal thickening in the right knee.
Because of the protracted course despite appropriate antimicrobial therapy, the patient's immune
system was investigated. The presence of neutrophils
in the infected synovial fluid was considered to reflect
adequate neutrophil chemotaxis and migration, so
laboratory tests of these functions were not performed. A nitroblue tetrazolium test, serum CH50,
and quantitative immunoglobulin values were all normal. Renal ultrasonography did not reveal abscess
formation or anatomic abnormalities predisposing
to renal infection. A barium enema showed normal
The isolate of enterococcus was examined for
antibiotic susceptibility using the tube dilution method
(3). The organism was sensitive to ampicillin with
minimal inhibitory and bactericidal concentrations of
0.625 pg/ml. One hour after ampicillin and gentamicin
infusions, serum was bactericidal for the patient's
enterococcus at a dilution of 1 : 1,024. Tests for synovia1 fluid bactericidal levels were not performed because
of the well-documented penetration of penicillins at
bactericidal concentrations into synovial fluid.
The patient was treated with gentamicin for a
total of 4 weeks and ampicillin for 6 weeks. The open
knee was closed secondarily, a draining sinus was left,
and the patient was discharged to a skilled nursing
facility, ambulating with a walker. One month later, he
was afebrile. The knee was slightly warm and drained
clear synovial fluid. Radiographs showed no progression of changes.
The patient was readmitted 2 months later with
an obvious posterior dislocation of his right tibia. At
this time the knee drained purulent fluid. The patient
was febrile (37.8"C), his peripheral white count was
1 1,400 with 75% polymorphonuclear leukocytes, the
erythrocyte sedimentation rate was 102 mm/hour, and
the serum alkaline phosphatase level was elevated to
346 IU. Radiographs revealed no evidence of osteomyelitis (Figure 2), but chronic pyogenic arthritis with
probable associated osteomyelitis was suspected. An
above-the-knee amputation was performed without
incident, and careful examination of the affected knee
confirmed chronic arthritis and osteomyelitis.
Figure 2. A and B, Patient I . radiographs of right knee on last admission, showing posterior dislocation and periosteal
Patient 2. A 61-year-old man with a 25-year
history of erosive rheumatoid arthritis was admitted
for fitting of a right lower extremity prosthesis. A right
ankle arthroplasty in 1975 had been followed by Morganellu rnorgani osteomyelitis necessitating removal
of the prosthesis, then amputation. A left total hip
arthroplasty done in 1976 had recently become painful.
Current antirheumatic therapy included D-penicillamine and prednisone 5 mg twice daily. Past complications of the patient’s arthritis included vasculitis and
pericarditis with tamponade.
On admission, vital signs were normal. Multiple
fixed deformities of rheumatoid arthritis were present
without inflammation. Scars of healed fingertip ulcers
and left foot sensory loss in stocking distribution were
present. The right below-the-knee amputation stump
was well-healed. The prosthetic left hip was painful at
all extremes of motion.
Admission urinalysis showed occasional white
blood cells; a culture was not performed. Radiographs
of the left hip showed loosening of the femoral component of the prosthesis, and aspiration of the left hip
under fluoroscopy produced fluid with 142,000 white
blood cells per ml, 98% of which were polymorphonuclear leukocytes. Cultures of the fluid yielded a pure
growth of enterococcus, sensitive to ampicillin at
minimal inhibitory and bactericidal concentrations of
1.0 pg/ml.
Therapy was begun with intravenous ampicillin
12 gm/day. Six days latcr tobramycin sulfate was
added and the hip prosthesis was removed. Serum
tobramycin levels were within accepted therapeutic
ranges. Cultures of the prosthesis bed obtained at
surgery were sterile, although pus was seen at the time
of capsulotomy. The patient received 2 weeks of
tobramycin sulfate therapy and 6 weeks of ampicillin.
His left hip, which underwent Girdlestone excision
arthroplasty, is sufficiently functional to allow limited
ambulation with a walker, despite the right below-theknee prosthesis.
Studies of this patient’s immune and leukocyte
function were not believed to be indicated; the prosthetic joints were thought to be the factors which
placed the patient at risk for developing joint infections. An excretory urogram demonstrated necrosis of
a single papilla in the right kidney and a small radiopaque stone in that kidney. Cystometrography
showed normal results.
Enterococci comprise a subset of group D
streptococci distinguished by growth in bile and 6.5%
NaC1, and relative insensitivity to penicillin G in usual
doses (1). Although enterococci are recognized as
important pathogens in urinary tract infections, endocarditis, cholecystitis, and cholangitis, they are rarely
isolated from synovial fluid. We are aware of only 5
well-documented cases of enterococcal arthritis reported since antibiotics have bccn in common use.
Hutto and Ayoub (4)reported a 6-week-old girl with
right humeral osteomyelitis and adjacent septic glenohumeral arthritis due to group B streptococcus and
simultaneous enterococcal arthritis of the right hip.
She did well following 4 weeks of intravenous ampicillin therapy and surgical drainage. Deshayes et a1 ( 5 )
described polyarticular mixed enterococcal and “non-
hemolytic streptococcal” infections in a 4dyear-old
butcher following a work accident. The patient was
treated with Bicillin and prednisone for 6 months
followed by therapy with oral ampicillin, chloramphenicol, and surgical drainage. Amputation of a leg
was cventually required to control infection. Three
additional cases of enterococcal arthritis were tabulated without description in 2 large series of cases (6,7)
which suggests that its occurrence is rare.
Most bacterial arthritis is thought to arise from
hematogenous seeding of synovium (2). Therefore, the
rarity of enterococcal arthritis is noteworthy in view of
the frequency of enterococcal bacteremia. Enterococci were the streptococci most frequently cultured from
blood during a 2-year period in the mid 1960s at the
Massachusetts General Hospital (8). The Centers for
Disease Control reported group D streptococci isofated from 6.2% of the nosocomial bacteremias included
in a 1978 study (9). It is also worth emphasizing the
frequency of nonarticular enterococcal infections
which probably place large numbers of patients at risk
for transient undocumented cnterococcal bacteremias.
Group D streptococci were isolated from 13.8% of
nosocomiai urinary tract infections (9), and enterococci have been cultured from bile in 5% of cholecystectomy specimens (10). Our first patient’s admission urinalysis and urine culture suggest a urinary tract source
of enterococcus which subsequently seeded the synovium. The source of enterococcus in the second patient is unknown.
Enterococcal bacteremia without other predisposing factors does not appear to be sufficient to
establish septic arthritis. A number offactors including
alcoholism ( I I ) , systemic glucocorticoid therapy (121,
osteoarthritis (13), rheumatoid arthritis (14). and prosthetic joints (15) have been reported to predispose
patients to septic arthritis. Although evidence supporting these reports is not always convincing and pathogenetic mechanisms are speculative, in this report
patient I abused alcohol and received high-dose glucocorticoids and patient 2 had 3 risk hctors (glucocorticoid therapy, rheumatoid arthritis, and prosthetic
The limited information available does, however, suggest a guarded prognosis for this form of
infectious arthritis. Six weeks of high-dose parenteral
antibiotics plus aggressive surgical debridement did
not prevent establishment of a chronic pyogenic arthritis and osteomyelitis in our first patient. Deshayes’
patient also underwent amputatlon after failurc of
prolonged antimicrobial therapy and repeated surgical
drainage. The second patient in the current report did
well after removal of his infected prosthesis and administration of antibiotic therapy, but an extended
period of observation will be needed to determine if he
is cured. Early diagnosis and institution of optimal
antimicrobial therapy may improve these suboptimal
results. Enterococci should be included in the differential diagnosis of infectious arthritis in any patient with
a preexisting infection at a site likely to involve
enterococci (cholecystitis, nosocomial urinary tract
infection, etc.), or with a recent history of genitourinary manipulation, or other conditions predisposing to
enterococcal bacteremia. Initial empiric therapy for
these patients should include agents effective against
the enterococcus, particularly if gram-positive cocci
are identified in Gram stained synovial fluid.
Our limited experience also suggests that confirmed cases of enterococcal arthritis should be treated
aggressively with a regimen appropriate for enterococcal endocarditis, such as 4 weeks of parenteral ampicillin with at least 2 weeks of aminoglycoside therapy
(16). It is hoped that greater awareness of this uncommon but destructive infectious arthritis will lead to
more effective treatment.
1 . Shufman JA: Streptococcus faecalis and other entero-
cocci, Principles and Practice of Infectious Diseases.
Vol. 1. Edited by GL Mandell. RG Douglas, JE Bennett.
New York, John Wiley and Sons, 1979, pp 1605-1609
Smith JW: Infectious arthritis, Principles and Practice of
Infectious Diseases. Vol. 1 . Edited by GL Mandell, RG
Douglas. JE Bennett. New York, John Wiley and Sons,
1979, pp 933-945
Gavan TL, Barry AL: Microdilution test procedures,
Manual of Clinical Microbiology. Third edition. Edited
by EH Lennette, A Balows, WJ Hausler, JP Truant.
Washington, American Society of Microbiology. 1980
Hutto JA, Ayoub EM: Streptococcal osteomyclitis and
arthritis in a neonate. Am J Dis Child 129:1449-1451,
Deshayes P, Schrub J, Adrian P: Osteo-arthritis rnultiples a cnterocoque favorisees par la corticotherapie.
Rev Rhum Ma\ Osteoartic 35:209-212, 1968
Sharp JT, Lidsky MD, Dufly J, Duncan MW: Infectious
arthritis. Arch Intern Mcd 139:1125-1130, 1979
tioldenberg DL, Brandt KD, Cohen AS, Cathcart ES:
Treatment of septic arthritis: comparison of needle
aspiration and surgery as initial modes of joint drainage.
Arthritis Rheum 18:83-90, 1975
Duma RJ. Weinberg A N , Medrek TF, Kuntz LJ: Streptococcal infections: a bacteriologic and clinical study of
streptococcal bacteremia. Medicine (Baltimore) 48:87127. 1969
9. Centers for Disease Control: National Nosocomial Infections Study Kcport, Annual Summary 1978, March
10. Jarvinen H, Kenkonen OV, Palmu A: Antibiotics in
acute cholecystitis. Ann Clin Res 10:247-251, 1978
I I . Willkens RE',Hcaley LA, Decker JL: Acute infectious
arthritis in the aged and chronically ill. Arch Intern Med
106:354-364, 1960
12. Mills LC, Boylston BF. Greene JA, Moyer JH: Septic
arthritis as a complication of orally given steroid therapy. JAMA 164:1310-1314, 1957
Ward JR, Atcheson SG: Infectious arthritis. Med Clin
North Am 61 :3 13-329, 1977
Karten I: Septic arthritis complicating rheumatoid arthritis. Ann Intern Med 70: 1147-1 158, 1%9
Eftekhan NS: Principles of Total Hip Arthroplasty.
Saint Louis. CV Mosby Company, 1978, p 552
Koenig MG, Kaye D: Enterococcal endocarditis: report
of nineteen cases with long term follow-up data. N Engl
J Med 264257-264, 1961
Без категории
Размер файла
385 Кб
septic, causes, arthritis, enterococcus, unusual
Пожаловаться на содержимое документа