See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/266148942

Septic Arthritis

Article  in  Hospital Medicine Clinics · October 2014

DOI: 10.1016/j.ehmc.2014.06.009

CITATIONS READS

0 6,458

3 authors, including:

Hyung J Cho
New York City Health and Hospitals Corporation
28 PUBLICATIONS   55 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Hyung J Cho on 27 October 2017.

The user has requested enhancement of the downloaded file.

S eptic A r t h r i t i s
a, b c
Hyung J. Cho, MD *, Leah A. Burke, MD , Mikyung Lee, MD

KEYWORDS
 Septic arthritis  Bacterial arthritis  Infectious arthritis

HOSPITAL MEDICINE CLINICS CHECKLIST

1. Understand the common sources of infection of septic arthritis.

2. Use the necessary diagnostic testing for septic arthritis.
3. Interpret synovial fluid analysis accurately.
4. Use the correct initial antibiotic regimen, based on clinical picture, synovial fluid
analysis, and Gram stain.
5. Treat with the appropriate duration and modality of antibiotics.
6. Understand the morbidity and mortality related with this disease.

DEFINITION

What is the definition of septic arthritis?

Septic arthritis historically refers to a bacterial infection in a joint, and can include other
causative agents such as fungal and mycobacterial.1

EPIDEMIOLOGY

What is the prevalence of septic arthritis in adults?

There is an estimate of 20,000 cases of septic arthritis per year in the United States.2
The incidence appears to be increasing, which may be accounted for by the increasing
use of immunosuppressive treatments and invasive procedures, in addition to an
aging population. The prevalence of bacterial arthritis presenting with acute monoar-
thritis ranges from 8% to 27%, as determined from prospective studies in Boston and
Taiwan.3,4

a
Division of Hospital Medicine, Department of Medicine, Icahn School of Medicine at Mount
Sinai, One Gustave L. Levy Place, Box 1086, New York, NY 10029, USA; b Division of Infectious
Diseases, Department of Medicine, Weill Cornell Medical College, 1300 York Avenue, Box 125,
New York, NY 10021, USA; c Division of Infectious Diseases, Icahn School of Medicine at Mount
Sinai, Guggenheim Pavilion Floor 11th Floor East Room 378D, 1190 Fifth Avenue, New York, NY
10029, USA
* Corresponding author.
E-mail address: hyung.cho@mountsinai.org

Hosp Med Clin 3 (2014) 494–503

http://dx.doi.org/10.1016/j.ehmc.2014.06.009
2211-5943/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
 Septic Arthritis 495

PATHOGENESIS

What are the risk factors for septic arthritis in adults?

The most important risk factor for septic arthritis is abnormal joint architecture. This
correlation is seen in patients with rheumatoid arthritis and osteoarthritis. Diabetics with
Charcot arthropathy are also at higher risk.5 Although less commonly described,
crystal-induced arthropathy still remains an important predisposing factor.6 Other notable
factors include older age, low socioeconomic status, intravenous drug abuse, alcoholism,
diabetes, previous intra-articular corticosteroid injection, and cutaneous ulcers.7

What are the common sources of infection?

Infection is introduced hematogenously or by direct inoculation. Hematogenous

spread is most common, and is more likely to occur in elderly patients and those
with predisposing factors such as intravascular devices, urinary catheters, injection
drug use, and immunosuppression.8,9 Direct inoculation occurs through trauma, joint
surgery, or bite wounds.7

What are the most common pathogens associated with septic arthritis?
Bacterial causes of septic arthritis are mostly caused by Staphylococcus aureus,
including methicillin-resistant S aureus. Streptococci are also frequent causes, along
with gram-negative bacilli. The clinical presentation of specific pathogens is shown in
Table 1.

HISTORY AND EXAMINATION

What are the most common symptoms of septic arthritis?

Patients usually present with a 1- to 2-week history of erythematous, painful, swollen,

and hot joint(s) with a decreased range of motion. Large joint involvement is the most
common presentation, and the knee is involved in about 50% of cases. Other common
joints include hip, shoulder, ankle, elbow, and wrist. Multiple joints are involved in 15%
of patients, and are more likely to occur in those with rheumatoid arthritis or other con-
nective tissue disorders and severe sepsis.10 Fever is present in most cases, but chills
are unusual.11–13 Symptoms can be less pronounced in the elderly, the immunocom-
promised, and intravenous drug abusers.

What are the most important features to note in the physical examination?

The physical examination helps determine if the site of inflammation is intra-articular or

periarticular. If the intra-articular space is affected, the joint is naturally held in the po-
sition where maximal space is allowed (ie, in extension for the knee joint). Any active or
passive range of motion will cause pain and limitation. Periarticular abnormality in-
volves structures such as the skin and bursa.

DIAGNOSIS

Which diagnostic tests are needed to diagnose septic arthritis?

If septic arthritis is suspected, arthrocentesis should be performed on presentation.

Gram stain, culture, leukocyte count and differential, and crystal analysis should be
496 Cho et al

Table 1
Clinical presentation of specific pathogens

Clinical History Joint Involvement Pathogen

Soft-tissue skin Monoarticular, polyarticular Staphylococcus aureus,
infection Streptococcus
Sexually active Polyarticular Neisseria gonorrhoeae
Elderly, UTI, skin Monoarticular Gram-negative rods, including
breakdown enteric organisms
Intravenous drug use Sternoclavicular, Sacroiliac, Pseudomonas, S aureus
pubic symphysis
Gardening injuries Monoarticular: knee, hand, Pantoea agglomerans, Nocardia
wrist asteroides, Sporothrix schenckii
Rheumatoid arthritis Monoarticular S aureus
Anti-TNF therapy Monoarticular Salmonella, Listeria
Animal bites Small joints Oral flora including anaerobes,
Pasteurella multocida,
Capnocytophaga canimorsus
Southwestern USA, Knee Coccidioides immitis
Central and South
America, respiratory
symptoms
Tick bite, erythema Oligoarthritis, particularly Borrelia burgdorferi
migrans large joints

Abbreviations: TNF, tumor necrosis factor; UTI, urinary tract infection.

Adapted from Sharff KA, Richards EP, Townes JM. Clinical management of septic arthritis. Curr
Rheumatol Rep 2013;15(6):1–9.

obtained. Blood cultures should be drawn to diagnose bacteremia, given the preva-
lence of hematogenous spread in septic arthritis. Erythrocyte sedimentation rate
(ESR) and C-reactive protein (CRP) are also useful markers, and can aid in diagnosis,
but are more useful for monitoring the treatment response.
A synovial fluid white blood cell count of greater than 50,000 cells/mm3 and poly-
morphonuclear leukocyte count of greater than 90% make the diagnosis of septic
arthritis likely. The presence of crystals does not exclude septic arthritis, and both
are known to cause synovial fluid leukocytosis of this degree. Gram stain is positive
in 29% to 50% of cases.11 The definitive diagnosis for septic arthritis is growth of bac-
teria in synovial fluid culture.

How can synovial fluid analysis distinguish bacterial arthritis from other causes of
acute monoarthritis?

The various causes of arthritis based on synovial fluid analysis are shown in Table 2.

Which radiographic studies are helpful for the diagnosis of septic arthritis?
Imaging studies are often ordered and may be helpful for the diagnosis, but none are
definitive for septic arthritis. Plain radiographs can help detect fractures, chondrocal-
cinosis, or inflammatory arthritis, effusions, and osteomyelitis. Ultrasonography can
help in detecting effusions and also in guiding joint aspirations and drainage.
Computed tomography or fluoroscopic guidance also aids in identifying arthrocente-
sis in joints that are difficult to examine.
 Septic Arthritis 497

Table 2
Synovial fluid analysis

WBC Count PMN Cell

Diagnosis Color Appearance (per mm3) Count (%) Gram Stain
Normal Clear Transparent <200 <25 Negative
Noninflammatory Straw Translucent 200–2000 <25 Negative
Inflammatory Yellow Cloudy 2000–100,000 >50 Negative
(noninfectious)
Gonococcal Yellow Cloudy- 34,000–68,000 >75 Variable (<50%)
opaque
Bacterial Yellow- Opaque >50,000 >75 Positive
(nongonococcal) green (60%–80%)
Lyme disease Yellow Cloudy 3000–100,000 >50 Negative
(mean: 25,000)

Abbreviations: PMN, polymorphonuclear; WBC, white blood cell.

Adapted from Horowitz DL, Katzap E, Horowitz S, et al. Approach to septic arthritis. Am Fam
Physician 2001;84(6):653–60.

MANAGEMENT

Which initial antibiotic regimens are preferred for septic arthritis?

Before starting treatment with antibiotics, a synovial fluid sample and 2 sets of blood
cultures should be obtained. Thereafter, empiric antibiotic therapy should be initiated
as soon as possible. There are no randomized controlled studies that have evaluated
antibiotic regimens for septic arthritis. The initial choice of antimicrobial regimens is
based on the clinical presentation and coverage of the most likely organisms to
cause infection, in addition to the synovial fluid Gram stain and culture. In general,
if the Gram stain of the synovial fluid shows gram-positive cocci, treatment with van-
comycin should be initiated. If the initial Gram stain shows gram-negative bacilli or
gram-negative cocci (the latter raises concern for Neisseria gonorrhoeae), therapy
with a third-generation cephalosporin (ceftriaxone, cefotaxime) should be started.
Ceftazidime or cefepime should be given when Pseudomonas aeruginosa is a sus-
pected pathogen. In patients with b-lactam allergy, aztreonam or ciprofloxacin
can be used. If the initial Gram stain is negative, it is reasonable to initiate therapy
with vancomycin in immunocompetent patients, and vancomycin plus a third-
generation cephalosporin in immunocompromised patients or those at risk for a
sexually transmitted disease. Modifications to the initial regimen should be made af-
ter synovial fluid culture and susceptibility results are available. A summary of sug-
gested empiric antibiotic treatment is presented in Fig. 1. A recommendation
regarding the choice of antibiotic therapy once the causative organism is identified
is presented in Table 3.

Are intra-articular antibiotics preferred over systemic antibiotics?

Intra-articular antibiotics are not recommended because effective parenteral or
oral therapy produces adequate levels of antimicrobial agents in joint fluid. In addi-
tion, direct instillation of antibiotics into a joint may cause an inflammatory
response.14
498 Cho et al

Positive Gram stain

Yes No

Gram positive cocci Vancomycin + third

generation cephalosporin

Clusters:
S. aureus: Nafcillin or oxacillin or
cefazolin 1-2g IV q8
MRSA suspected: Vancomycin
15mg/kg IV q12
Pairs and chains:
Streptococci: Penicillin G does to:
20-24 million units/d by continuous
infusion or divided into 6 doses (i.e. q4h)

Gram-negative diplococcic

Suspected: N. gonorrhoeae or
N. meningitides: Ceftriaxone 1-2g/24h

Gram-negative bacilli

Ceftriaxone 2g/24h or
ceftazidime 2g/8h

Fig. 1. Algorithm for initial empiric antibiotic therapy for bacterial arthritis. (Modified from
Tomsic M, Praprotnik S, Louie JS, et al. Septic arthritis. In: Weisman MH, ed. Targeted treat-
ment of the rheumatic diseases. Philadelphia, PA: Elsevier; 2010; with permission.)

What is the appropriate duration of systemic antibiotics?

There have been no controlled trials examining the duration of antimicrobial therapy in
bacterial arthritis, and treatment recommendations are based on clinical case series.
A 2007 systematic review published by Mathews and colleagues15 reported that “little
good quality evidence exists to guide the diagnosis and management of septic arthritis.
Key unanswered questions remain surrounding the medical and surgical management of
the infected joint.” The investigators report that “antibiotics are clearly required for a pro-
longed period, but there are no data to indicate by which route or for how long.”15
Guidelines from the United Kingdom recommend parenteral therapy for 2 weeks fol-
lowed by 4 weeks of oral therapy.16,17 Gonococcal joint infections can be treated with
a 2-week course of ceftriaxone.18
Clinical practice guidelines from the Infectious Diseases Society of America sup-
port a 4-week course of parenteral therapy in patients with infectious arthritis caused
 Septic Arthritis 499

Table 3
Choice of antibiotics for treatment of infectious arthritis based on causative organism

Organism Antibiotic of First Choice Alternative Antibiotic

S aureus, methicillin- Vancomycin Linezolid, daptomycin
resistant
S aureus, methicillin- Nafcillin Cefazolin, vancomycin
susceptible
Coagulase-negative Vancomycin Linezolid, daptomycin
Staphylococcus
Group A Streptococcus, Penicillin Cefazolin, ceftriaxone, clindamycin
S pyogenes
Group B Streptococcus, Penicillin Cefazolin, ceftriaxone, clindamycin
S agalactiae
Enterococcus spp Ampicillin (if susceptible) Linezolid, daptomycin
or vancomycin
N gonorrhoeae or Ceftriaxone Quinolones (ciprofloxacin or
N meningitidis levofloxacin)
Gram negatives (other Ceftriaxone or Aztreonam, quinolones (ciprofloxacin
than Neisseria) cefotaxime or levofloxacin)
Pseudomonas Cefepime or ceftazidime Piperacillin-tazobactam, carbapenems
aeruginosa (imipenem, meropenem, or
doripenem); Significant b-lactam
allergy: quinolones (ciprofloxacin,
levofloxacin), aztreonam

Data from Ohl CA. Infectious arthritis of native joints. In: Mandell G, Bennett J, Dolin R et al, edi-
tors. Mandell, Douglas, and Bennett’s principles and practice of infectious diseases. 7th edition.
Philadelphia: Churchill Livingstone; 2010. p. 1443–56; and Osmon DR, Berbari EF, Berendt AR,
et al. Diagnosis and management of prosthetic joint infection: clinical practice guidelines by the
Infectious Diseases Society of America. Clin Infect Dis 2013;56:e1–25.

by S aureus and gram-negative bacilli.19 Patients with prosthetic joint infections

should been seen in consultation with orthopedic surgeons and infectious diseases
specialists to determine whether the prosthetic material can be retained, and
whether the patient requires suppressive antibiotic therapy if such hardware is not
removed.20

When is joint drainage indicated, and what is the optimal method?

Removal of bacteria and inflammatory debris from the joint is an essential component
of the management of infectious arthritis. The most effective method of drainage has
yet to be clearly determined, given that there have been no randomized controlled
studies regarding joint-drainage procedures for adults in this setting. Joint drainage
should be performed as early as possible, because the infection occurs in a closed
space similarly to an abscess, and inflammatory cells release cytokines and proteases
that cause cartilage degradation and inhibit cartilage synthesis. In addition, bacterial
DNA and bacterial toxins are also believed to have a deleterious effect on joint
structures.21
Any of the 3 drainage procedures may be used: closed needle aspiration, arthro-
scopic drainage, or arthrotomy (open surgical drainage). Most peripheral joints can
be drained with closed needle aspiration. This method has historically been used in
500 Cho et al

less severe cases and in distal, smaller joints. It is less invasive than surgical drainage
and may be associated with more rapid functional recovery, but has not been associ-
ated with shorter length of stay in hospital or decreased mortality.22 If adequate
drainage cannot be obtained by needle aspiration, either arthroscopy or open
drainage is necessary. Arthroscopy or open drainage is often preferred for knee,
shoulder, axial, and prosthetic joints, owing to better visualization of the joint space
and easier irrigation and drainage.23
After initial joint drainage, response to therapy should be monitored with serial sy-
novial fluid analyses (which should demonstrate that the fluid has become sterile
and the total leukocyte count is decreasing). If this is not the case, repeat drainage
and/or a change to the antimicrobial regimen is necessary. One may also be reassured
by a decrease in serum inflammatory markers, such as ESR and CRP, as a sign of a
good response to therapy.

PROGNOSIS

What is the prognosis after septic arthritis is diagnosed?

The prognosis of septic arthritis is related to host factors, such as prior joint damage,
the virulence of the infecting organism, and the speed with which adequate treatment
is begun. A prospective community survey assessed the outcome of bacterial arthritis
in Amsterdam between October 1990 and October 1993. Of 150 patients with a total of
174 infected joints, overall health outcomes were poor in 21%, predominantly in pa-
tients with preexisting joint disease. Fourteen patients (10%) died during the course
of bacterial arthritis. Adults with preexisting joint disease and/or the presence of
infected joints with synthetic material stayed in hospital longer and underwent more
joint surgery than adults without preexisting joint disease. After discharge from the
hospital, surviving patients with preexisting joint disease had more physiotherapy
and longer admissions to rehabilitation clinics or nursing homes. Nearly half of the
infected joints in adults had a poor joint outcome (as defined by the need for amputa-
tion, arthrodesis, prosthetic surgery, or severe functional deterioration), regardless of
whether there was preexisting joint disease. Adverse prognostic factors included older
age, preexisting joint disease, and an infected joint containing synthetic material.24
Certain bacterial pathogens also confer a worse prognosis after treatment. For
example, in patients with pneumococcal bacterial arthritis, 95% of adults had a return
to baseline joint function or only mild limitation of joint motion following the completion
of therapy.25,26 On the other hand, studies of S aureus bacterial arthritis report that
46% to 50% have poor joint outcomes following therapy.16

CLINICAL GUIDELINES

BSR1, BHPR2, BOA3, RCGP4, and BSAC5 guidelines for management of the hot
swollen joint in adults. Rheumatology 2006;45:1039–41.
Infectious Diseases Society of America. Diagnosis and management of prosthetic
joint infection: clinical practice guidelines by the Infectious Diseases Society of Amer-
ica. Clin Infect Dis 2013;56:e1–25.

1
British Society for Rheumatology.
2
British Health Professionals in Rheumatology.
3
British Orthopaedic Association.
4
Royal College of General Practitioners.
5
British Society for Antimicrobial Chemotherapy.
 Septic Arthritis 501

PERFORMANCE IMPROVEMENT

How can the risk of septic arthritis be decreased in hospitalized patients?

The risk of septic arthritis can be reduced in the hospital setting via the following ac-
tions on the part of the health care team:
1. Prompt treatment of local skin infections (including cellulitis and cutaneous ulcers),
especially those that overlie joint surfaces.
2. Prompt identification and control of bacteremia.
3. Removal of intravenous catheters as soon as possible, and only placing such cath-
eters when needed.
4. Exercising discretion regarding the use of intra-articular corticosteroid injections
and invasive orthopedic procedures in patients at high risk for septic arthritis. If
there is doubt about whether infection might be present when evaluating a hot,
swollen, erythematous joint, intra-articular steroids should not be used.

What can be done to alter morbidity and mortality associated with septic arthritis?

The authors recommend the following actions to reduce the potential morbidity and
mortality from septic arthritis:
1. Joint aspiration for a synovial fluid sample before initiation of antibiotics.
2. Initiation of empiric antibiotic therapy as soon as possible, guided by the most likely
infecting organisms and the result of the synovial fluid Gram stain.
3. Adjustment of antibiotic therapy as soon as synovial fluid culture and susceptibility
results are available.
4. Prompt drainage of the infected joint, with involvement of orthopedic surgeons and
infectious diseases specialists if arthroscopy or open drainage is warranted and/or
prosthetic joints are involved.
5. Monitoring response to therapy with serial synovial fluid analyses with or without
serum inflammatory markers (ESR, CRP).
6. Treatment with antibiotics for at least 3 to 4 weeks, with initial parental therapy.
The British guidelines regarding the management of the hot swollen joint in adults
recommend a mechanism for audit in cases of proven septic arthritis as a means to
improve patient outcomes17:
1. Was the joint aspirated at presentation before antibiotics? If not, what was the
reason?
2. Was there a delay in treatment and, if so, why?
3. Was ESR and CRP measured at diagnosis and serially?
4. Were appropriate cultures taken?
5. Was the initial antibiotic choice in keeping with the guidelines?
6. Was prosthetic joint sepsis managed by orthopedic surgeons?
Because there are currently no formal national guidelines available to guide the gen-
eral management of septic arthritis in adults without prosthetic joints, the authors
advocate the development of local institutional guidelines for managing this condition,
in addition to close monitoring of compliance with these guidelines.

REFERENCES

1. Newman JH. Review of septic arthritis throughout the antibiotic era. Ann Rheum
Dis 1976;35(3):198–205.
502 Cho et al

2. Sharff KA, Richards EP, Townes JM. Clinical management of septic arthritis. Curr
Rheumatol Rep 2013;15(6):1–9.
3. Shmerling RH, Delbanco TL, Tosteson AN, et al. Synovial fluid tests: what should
be ordered? JAMA 1990;264(8):1009–14.
4. Jeng GW, Wang CR, Liu ST, et al. Measurement of synovial tumor necrosis factor-
alpha in diagnosing emergency patients with bacterial arthritis. Am J Emerg Med
1997;15(7):626–9.
5. Kaandorp CJ, Schaardenburg DV, Krijnen P, et al. Risk factors for septic arthritis
in patients with joint disease. Arthritis Rheum 1995;38(12):1819–25.
6. Yu KH, Luo SF, Liou LB, et al. Concomitant septic and gouty arthritis—an analysis
of 30 cases. Rheumatology 2003;42(9):1062–6.
7. Mathews CJ, Weston VC, Jones A, et al. Bacterial septic arthritis in adults. Lancet
2010;375(9717):846–55.
8. Gavet F, Tournadre A, Soubrier M, et al. Septic arthritis in patients aged 80 and
older: a comparison with younger adults. J Am Geriatr Soc 2005;53(7):1210–3.
9. Luca-Harari B, Ekelund K, van der Linden M, et al. Clinical and epidemiological
aspects of invasive Streptococcus pyogenes infections in Denmark during 2003
and 2004. J Clin Microbiol 2008;46(1):79–86.
10. Abelson A. Septic arthritis. Available at: Clevelandclinicmeded.com. Accessed
August 1, 2010.
11. Margaretten ME, Kohlwes J, Moore D, et al. Does this adult patient have septic
arthritis? JAMA 2007;297(13):1478–88.
12. Horowitz DL, Katzap E, Horowitz S, et al. Approach to septic arthritis. Am Fam
Physician 2001;84(6):653–60.
13. Abelson A. Septic arthritis. In: Post TW, editor. UpToDate. Waltham (MA): UpTo-
Date. Available at: UpToDate.com. Accessed August 1, 2010.
14. Goldenberg DL, Reed JI. Bacterial arthritis. N Engl J Med 1985;312:764.
15. Mathews CJ, Kingsley G, Field M, et al. Management of septic arthritis: a system-
atic review. Ann Rheum Dis 2007;66(4):440–5.
16. Weston VC, Jones AC, Bradbury N, et al. Clinical features and outcome of
septic arthritis in a single UK Health District 1982–1991. Ann Rheum Dis
1999;58:214–9.
17. Coakley G, Matthews C, Field M, et al, British Society for Rheumatology Stan-
dards, Guidelines and Audit Working Group. BSR & BHPR, BOA, RCGP and
BSAC guidelines for management of the hot swollen joint in adults. Rheumatology
2006;45:1039–41.
18. Ohl CA. Infectious arthritis of native joints. In: Mandell G, Bennett J, Dolin R, et al,
editors. Mandell, Douglas, and Bennett’s principles and practice of infectious dis-
eases. 7th edition. Philadelphia: Churchill Livingstone; 2010. p. 1443–56.
19. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the Infectious
Diseases Society of America for the treatment of methicillin-resistant Staphylo-
coccus aureus infections in adults and children: executive summary. Clin Infect
Dis 2011;52:285–92.
20. Osmon DR, Berbari EF, Berendt AR, et al. Diagnosis and management of pros-
thetic joint infection: clinical practice guidelines by the Infectious Diseases Soci-
ety of America. Clin Infect Dis 2013;56:e1–25.
21. Deng GM, Tarkowski A. The features of arthritis induced by CpG motifs in bacte-
rial DNA. Arthritis Rheum 2000;43:356.
22. Ravindran V, Logan I, Bourke BE. Medical vs surgical treatment for the native joint
in septic arthritis: a 6-year, single UK academic centre experience. Rheuma-
tology 2009;48:1320–2.
 Septic Arthritis 503

23. Stutz G, Kuster MS, Kleinstück F, et al. Arthroscopic management of septic

arthritis: stages of infection and results. Knee Surg Sports Traumatol Arthrosc
2000;8:270.
24. Kaandorp CJ, Krijnen P, Moens HJ, et al. The outcome of bacterial arthritis – a
prospective community-based study. Arthritis Rheum 1997;40(5):884–92.
25. Ross JJ, Saltzman CL, Carling P, et al. Pneumococcal septic arthritis: review of
190 cases. Clin Infect Dis 2003;36:319.
26. Shirtliff ME, Mader JT. Acute septic arthritis. Clin Microbiol Rev 2002;15(4):
527–44.

View publication stats