• 
• 

Three Part Question

In [adults with suspected arthritis] is [a negative Gram stain of the joint aspirate ]sufficient [to rule out septic arthritis]

Clinical Scenario

A 30 year old Nurse presents to the ED with two day history of a hot, swollen joint. She reports no prior knee swelling and no recent trauma or knee surgery, illegal drug use, rash, uvetis or risky sexual behavior. On examination, she is afebrile and has a left knee effusion. Her WBC, ESR and CRP are all raised. The joint aspirate reports a negative gram stain. You wonder if negative gram stain is enough to rule out septic arthritis?

Search Strategy

Multifile search using OVID – Medline (1950-2008), Embase (1980-2008), CINHAL (1982-2008), Cochrane ({[exp.Infectious Arthritis OR hot joint.mp OR septic arthritis.mp OR septic joint.mp] AND [exp. Gram stain OR gram stain$.mp] AND [exp Synovial Fluid OR aspirate$.mp OR synovialfluid$.mp]}) Limited to Humans, English Language and Adults

Search Outcome

The following number of articles were identified from each of the databases: Medline 20 citations, Embase 22 citations, CINAL 0 citations, Cochrane 1 citations. Duplicates and irrelevant titles removed from 43 articles leaving a total of 8 relevant articles.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Daniel C, et al 2007 USA	49 culture positive synovial fluid aspirates in an ED in an urban tertiary care center	Case Series (Level 3)	Gram stain	Negative in 27/49 (55%;95%CI, 37-74%)	Imperfect Gold standard*
Nausha A, et al. 2006 Pakistan	116 patients from a tertiary care hospital identified by the ICD 711 as having SA	Retrospective Cohort (Level 3)	Gram Stain	78/116(67.2%) had a gram stain done 26/78 (33.3%) stained positive for a specific organism and 52/78 (66.7%) inconclusive	Small sample. Narrow definition of SA.
Faraj A, et al 2002 UK	22 patients of who had arthroscopic lavage following a provisional diagnosis of SA who later were all found to have culture proven SA	Case Series (Level 3)	Gram Stain	Sensitivity 45% Specificity 100%	Case Series. Small sample. No confidence intervals
Margaretten ME, et al. 2008 USA	Fourteen studies involving 6242 patients, of whom 653 diagnosed with SA and met all respective inclusion criteria.	Systematic Review (Level 1)	Gram Stain	Seven studies identified that reported sensitivities of 29-50%[1-5]	No confidence intervals. Imperfect gold standard*
McCutchan, et al. 1989 USA	41 patients diagnosed with SA on a positive cultures	Case Series(Level 3)	Gram Stain	Positive	Case Series. Small Sample

Comment(s)

There is no evidence to suggest that a gram stain is sensitive enough to rule out septic arthritis such that a negative gram stain should not be taken as evidence to rule out SA.[1-8] This is supported by studies that have noted the absence of gram stain with a positive culture.[Mc Cutchan, Daniel] Most of the studies of the value of gram stain have been case series, therefore specificity is often not calculated. In the one retrospective study that looked at patients undergoing arthroscopic lavage following a provisional diagnosis of SA, specificity was calculated as 100%. [Faraj] A positive gram stain can therefore rule in (Spin) SA.

Editor Comment

RB

Clinical Bottom Line

A negative gram stain is not sufficiently sensitive to rule out septic arthritis. A positive gram stain is sufficient to rule in SA. *Imperfect Gold Standard –gram stain (sensitivity 29-50%) and culture (sensitive-82%)[Margaretten].

References

1. Daniel C, et al. How sensitive is the synovial fluid white blood cell count in diagnosing septic arthritis? American Journal of Emergency Medicine. 2007;749-752
2. Nausha A, et al Septic arthritis in a tertiary care hospital Journal of Pakistan Medical Association
3. Faraj A, et al. Gram staining in the diagnosis of septic arthritis Acia Orthopadica Belgica 2002;388-391
4. Margaretten et al. Does the Adult Patient have Septic Arthritis? JAMA 2008;1478-1488
5. McCutchan HJ, et al. Synovial leukocytosis in infectious arthritis. Clinical Orthopaedics and Related Research . 2002;388-391
6. Soderquist B. Bacterial or Crystal –associated Arthritis? Discriminating Ability of Serum Inflammatory Markers. Scandinavian Journal of Infectious Diseases. 1998;30:591-596
7. Krey PR. Synovial fluid leukocytosis :a study of extremes. American Journal of Medicine. 1979;67:436-442.
8. Korekangas P. Synovial fluid in bacterial arthritis vs. reactive arthritis and rheumatoid arthritis in the adult knee. Scandinavian Journal of Rheumatology. 1992;21:283-2886.