Laboratory Rounds Is this a Septic Joint? Mark Boyko, R3 EM.

Laboratory RoundsIs this a Septic Joint?

Mark Boyko, R3 EM

Case

53 yo female comes in with 2 day history of increasing R knee pain, now giving her a limp. Does not recall injuring it

That knee is always ‘a little sore’ from running injuries years ago

Case

PMHx: HTN GERD Smoker Gout (toes, L ankle) – hasn’t had a flare in years

Meds Allopurinol Ramipril Ranitidine

Case

Phx No fever, normal vitals Knee looks swollen, no

cellultis Joint warm, ROM is

painful but patient can do it

Labs

Serum WBC 14 ESR 32 CRP 17 Uric Acid 400 Synovial Fluid

WBC 36 x109/L, PMN’s 65% Low glucose Negative for crystals

Labs

Gram Stain Negative

What do you want to do??

Overview

Value of serum labs Value of synovial fluid analysis Gram’s Stain & Cultures Prosethetic Joints Course of Action for Dry Taps

Review – The Swollen Joint

Non-inflammatory Trauma OA

Inflammatory RA Crystal arthropathies Seronegative arthropathies Septic joint

Why is this Important?

Joint destruction can occur within 2-3 days if untreated infection

Patients can become systemically septic from a joint infection rather easily

We need to make decisions before cultures come back

We Love Prediction Tools

Can anything help us rule this out??

Serum Labs

Serum WBC >10 x 109/L sensitivity of 50% for infection LR 1.4 Many sterile but inflammatory joints give

elevated serum WBC

Bottom Line: Not sensitive

Serum Labs

Serum ESR ‘Elevation’ in most studies >30 mm/h Sensitive but not specific LR 1.3

Bottom Line: Only useful to track resolution of the infection over time

Serum Labs

Serum CRP ‘Elevated’ in most studies >100 mg/L Sensitivity 75%, poor specificity LR 1.6

Bottom Line: Although CRP shows promise, there is insufficient evidence for its sensitivity to be high enough to rule out septic arthritis.-Best Bets 2008

Synovial Fluid

What’s Normal? Normal knee has avg 4cc synovial fluid Normal synovial WBC <0.2x109/L Glucose same as plasma Uric Acid same as plasma Protein <25% of plasma

Synovial Fluid

Normal – amber, transparent

Synovial Fluid

Inflammatory Cells - opaque

Synovial Fluid

Hemarthrosis

Hemarthrosis

Trauma #1 cause Anticoagulation therapy Hemophilia Synovioma

Rarely, infection and hemarthrosis co-exist. If concerned, send for culture.

Synovial Fluid

Findings Normal Non-Inflamm Inflammatory Septic

Colour Clear Yellow Yellow Yellow

Clarity Transparent Transparent Opaque Opaque

WBC (x109/L) <0.2 0.2 - 2 2 - 150 20 - 200

PMN’s <25% <25% >50% >75%

Synovial Fluid

Glucose and Protein Synovial / Serum Glucose <0.5-0.75, low

sensitivity Synovial Glucose <1.5 mmol/L sensitivity 38-64% Synovial Protein >25% of plasma, low sensitivity

Bottom Line: Glucose and Protein levels have no role in the work up of a septic joint

Synovial Fluid

LDH >250 U/L was 100% sensitive in retrospective study

on 8 confirmed cases, prospectively was not as strong

Lactic Acid 90-97% NPV, but low powered studies

Bottom Line: Insufficient data to date

Synovial Fluid

Tumour Necrosis Factor – α Jeng et al, Am J Emerg Med 1997 Prospective, n=75 Synovial TNF-α >36.2 pg/mL sens 95%, spec

50% for bacterial infection

Bottom Line: Needs more study before routine order

Synovial Fluid

WBC

Margeretten et al, JAMA 2007

<25 LR 0.32

>25 LR 2.9

>50 LR 7.7

>100 LR 28.0

Synovial Fluid

30% of immunocompetent people with culture confirmed septic joint have synovial WBC <50 - McGillicuddy et al, Am J Emerg Med. 2007

50% of immunocompromised people with culture confirmed joint infection had WBC <28-McCutchan et al, Clin Orthop Relat Res 1990

Synovial Fluid

PMN’s

Margeretten et al, JAMA 2007

<90% LR 0.34

>90% LR 3.4

Synovial Fluid

WBC Bottom Line Cut-off of 50 x109 /L too insensitive rule-out

infection Use in clinical context The diagnostic cut-off that maximized the

sensitivity / specificity was a synovial WBC count of 17.5 x109/L (Sens 83%, Spec 67%)

- Li et al, Emerg Med J 2007

Synovial Fluid

Eosinophilia Parasitic infection Allergy Fungal Neoplasm Lyme disease

Combined Value?

Li et al, Emerg Med J 2007 Retrospective chart review 156 patients Combined Sensitivity 100% if:

Serum WBC <11 Serum ESR <20 Synovial WBC <50

Bottom Line: Not powered enough, not prospectively validated, cannot use to rule out septic joint

Synovial Fluid

Crystals Gout - Monosodium Urate, 90% sensitive, LR 14 Pseudogout – PPDC, 80% sensitive, LR 2.6 Cholesterol crystals – seen in chronic

inflammatory conditions

Crystals & Infection

Crystals do not rule out infection!

Retrospective study n=265 patients with crystals, 1.5% had septic joint

-Shah et al, J of Emerg Med 2007

Literature ranges from 1-20% of infectious joints co-exist with crystals

Microbio Review

ALL AGES: #1 cause still Staph Aureus

<30, sexually active: Neisseria Gonorrhea

Elderly: Gram Negatives

Prosthetics: Careful of Pseudomonas

Gram’s Stain

Guides your antibiotic therapy while awaiting cultures

Need roughly 3-5cc for stain & culture Only 65% sensitive for non-gonococcal infections Only 25% sensitive for gonococcal infections

Bottom Line: A negative Gram stain means nothing. A positive Gram stain means you should start treatment.

Cultures

‘Gold standard’ ? Gonococcus difficult to culture

Negative 50% of the time Requires chocolate agar

Non-gonococcus will culture 90% of time If you only have enough fluid for one test, this is

what you do Blood cultures reveal pathogen 25-50% of the

time

Gonococcal Arthritis

Synovial WBC often <50

Gram stain Positive only 25% of the time

Culture Positive only 50% of the time

If you suspect it, culture at 3 mucosal sites (pharynx, genitals, anus) will increase your chance of positive culture to 80%

Generally less destructive to the joint versus other pathogens

Gram Stain Positive, Culture Negative

In reality, this is retrospective

Go with your Gram Stain treat these patients while awaiting cultures

How does this happen? Antibiotics already on board Organism difficult to culture Was infected, now clearing

Prosthetic Joints <3mos since surgery

likely Staph Epiderm

>3mos since surgery Staph, Strep, Gram Neg

Should always call Ortho before tapping these in ER

Prosthetic Joints

Trampuz et al, Amer J of Med 2004

Prospective, n=133, 34 had septic joint Synovial WBC >1.7 x109/L , sens 94% spec 88% Synovial PMN’s >65%, sens 97% spec 98%

Mason et al, J of Arthroplasty 2003

Retrospective n=86 knees Ideal sensitivity 98% for synovial WBC 2.5 x109/L and

PMN’s 60%

What About Those Dry Taps?

Dry Tap?

Makes a septic joint unlikely usually a large enough effusion for tap, but never been validated

Options U/S guided in the department Consult Ortho Fluoroscopy guided

BOTTOM LINE: You need a sample of that fluid if you are worried about infection

Hot Joint, No Organism

Fastidious organism Antibiotics begun before cultures sent Wrong Diagnosis

Help increase your yield? Use blood culture bottles for synovial fluid

(aerobic and anaerobic)- Joint, Bone, Spine 2006

Relevance to Pediatrics?

No good studies specifically on synovial fluid analysis in the pediatric population

Most use numbers from adult data

How Many Use Kocher’s Criteria?

Kocher et al, J of Bone Joint Surg 2004

TAKE HOME MESSAGE

Cannot rely on serum values to rule out septic joint

If you believe there’s an effusion, get that fluid somehow

Unfortunately, nothing has a strong NPV

TAKE HOME MESSAGE

Synovial fluid: WBC & PMN is helpful WBC <18 is low risk but not zero WBC >50 is high likelihood PMN’S >90% is high likelihood Glucose, Protein useless

TAKE HOME MESSAGE

‘Gold Standard’ is clinical suspicion of an experienced physician, not laboratory tests (Current Opinion Rheumatology 2008)

Prosthetic Joints Lower WBC & PMN threshold

Don’t feel bad - 30% of the time reason for effusion remains ‘unknown’

Thanks

Feel free to ask for any references