Original Contribution

How sensitive is the synovial fluid white blood cell count in diagnosing Septic arthritis?

Daniel C. McGillicuddy MD^a,*, Kaushal H. Shah MD^b, Ryan P. Friedberg MD^a,

Larry A. Nathanson MD^a, Jonathan A. Edlow MD^a

^aDepartment of Emergency Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School,

Boston, MA 02215, USA

^bDepartment of Emergency Medicine, St. Luke’s Roosevelt Hospital, University Hospital of Columbia University, College of Physicians and Surgeons, New York, NY, USA

Received 22 May 2006; revised 7 December 2006; accepted 7 December 2006

Abstract

Objective: This study was conducted to determine the sensitivity of the current standard for synovial fluid leukocytosis analysis in diagnosing infectious arthritis or a septic joint. How accurate is the standard synovial fluid white blood cell (WBC) cutoff of 50000 WBC/mm³ to rule out septic arthritis? Methods: We conducted a retrospective study at an urban tertiary care medical center with 50000 adult emergency department visits per year. The study population consisted of patients with infectious arthritis confirmed by synovial fluid culture growth of a pathogenic organism. The study period lasted from January 1996 to December 2002. Extracted data included synovial fluid leukocyte count, Gram’s stain, culture, past medical history, and discharge diagnosis. Fisher exact test was used to compare proportions. Sensitivity and means were calculated with 95% confidence intervals (CI).

Results: There were 49 culture-positive synovial fluid aspirates in the 6-year study period. Nineteen (39%) of 49 patients (95% CI, 25%-52%) had a synovial WBC of less than 50000/mm³ and 30 (61%) of

49 patients (95% CI, 48%-75%) had a synovial WBC of more than 50000/mm³. The sensitivity of the

50000 synovial WBC/mm³ cutoff was 61% (95% CI, 48%-75%). Twenty-seven (55%) of 49 patients

had a negative Gram’s stain (95% CI, 41%-69%) and 15 (56%) of 27 patients (95% CI, 37%-74%) with negative Gram’s stain had a synovial WBC of less than 50000/mm³.

Conclusion: A synovial fluid WBC cutoff of 50000/mm³ lacks the sensitivity required to be clinically useful in ruling out infectious arthritis.

D 2007

Introduction

Atraumatic joint pain is a common complaint encoun- tered in the emergency department (ED). Because septic

* Corresponding author.

E-mail address: [email protected] (D.C. McGillicuddy).

arthritis is 1 of the most serious causes of atraumatic monoarticular joint pain, the emergency physician must make this diagnosis, usually by arthrocentesis and Synovial fluid analysis, in a timely manner.

The American Rheumatologic Association provides the following guidelines for synovial fluid cell count interpre- tation: noninflammatory N 200 to 2000 synovial fluid

0735-6757/$ - see front matter D 2007 doi:10.1016/j.ajem.2006.12.001

Table 1 Baseline demographics
Age (y) (range)	63 (17-93)
Sex (male)	51%
Elderly (age N65 y)	55%
Predisposing condition	84%

white blood cell count /mm³; inflammatory = 2000 to 50000 WBC/mm³; infectious = more than 50000 WBC/mm³ [1]. In the past 2 decades, several authors have reported instances of culture-proven septic arthritis with synovial WBC less than 50000/mm³ [2-5]. The question is how accurate is the standard synovial fluid WBC cutoff of less than 50000 WBC/mm³ to rule out septic arthritis?

We evaluated the sensitivity of the synovial WBC cutoff of 50 000/mm³ to diagnose septic arthritis at our institution.

Methods

Study design

This retrospective case study was conducted to deter- mine the sensitivity of the synovial WBC cutoff of 50000 WBC/mm³ to diagnose septic arthritis. The institutional review board at our hospital approved this study.

Study setting and population

The study population includes all patients with infectious arthritis confirmed by synovial fluid culture with growth of a pathologic organism at an academic, urban tertiary care center with 50000 ED patient visits per year. The study period lasted from January 1996 to December 2002.

Study protocol

Patients were included if they were older than 16 years and were diagnosed with septic arthritis by synovial fluid culture. The hospital medical records data base was queried via International Classification of Disease Ninth Revision (ICD-9) codes for septic arthritis or infectious arthritis, as well as arthrocentesis data. Next, all records were examined to determine if the ICD-9 diagnosis was consistent with the patient’s presentation, as well as confirmation of synovial fluid cell count and culture. All cultures were reviewed to evaluate for possible contaminants, and those that were listed as contaminants (eg, broth only growth) were

Table 2 Location of affected joints
Knee	32 (65%)
Hip	6 (12%)
Shoulder	3 (6%)
Elbow	3 (6%)
Wrist	2 (4%)
Ankle	2 (4%)
Sternoclavicular	1 (2%)

excluded. Demographics (age, sex), clinical (past medical history for predisposing risk factors, ie, history of joint surgery, end-stage renal disease, diabetes, cancer, immuno- suppression, history of previous septic joint, or history of arthritis.), and laboratory (synovial WBC, culture, and Gram’s stain) data were obtained from the patients’ electronic medical record. Fisher exact test was used to compare proportions. Sensitivity and means were reported with 95% confidence intervals (CI).

Table 3 Isolated pathogen from synovial fluid culture
Staphloccous aureus	27	55%
Beta streptococcus	11	22%
Escherichia coli	2	4%
Streptococcus pneumoniae	2	4%
Enterococcus	1	2%
Listeria	1	2%
Pseudomonas	1	2%
Nisseria gonorrhea	1	2%
Alpha streptococcus	1	2%
Seretria	1	2%
Bacillius subtillis	1	2%

Results

The database identified 49 patients with culture-positive synovial fluid aspirates in the 6-year study. Patient demo- graphics are summarized in Table 1. The average age was 63 years with roughly equal number of men and women.

The most common joint involved is the knee (65%), followed by the hip (12%) (Table 2). The most common organism involved is Staphylococcus aureus (55%) fol- lowed by beta streptococcus (22%) (Table 3).

Nineteen (39%) of 49 patients (95% CI, 25%-52%) had a synovial WBC of less than 50000/mm³ (Fig. 1). The sensitivity of the 50000 synovial WBC/mm³ cutoff was 61% (95% CI, 48%-75%). Twenty-seven (55%) of 49 patients had a negative Gram’s stain (95% CI, 41%-69%) and 15 (56%) of 27 patients (95% CI, 37%-74%) with negative Gram’s stain had a synovial WBC of less than 50000/ mm³.

Forty-one patients (84%; 95% CI, 73%-94%) had 1 or more predisposing medical conditions (history of joint surgery, end-stage renal disease, diabetes, cancer, immuno- suppression, history of previous septic joint, or history of arthritis). Comparison of specific medical conditions and synovial WBC levels showed no statistically signifi- cant relationship.

Discussion

Septic arthritis is a rare but serious cause of mono- articular joint pain that results from bacterial invasion of the synovium [6]. This invasion can result from direct

Fig. 1 Synovial WBC in adults with septic arthritis.

inoculation, hemotogenous spread, or spread of a local infection. According to Kaandorp et al [7], the incidence of septic arthritis has remained stable at approximately 5.7 of 100000 patients despite the advent of newer antibiotics and aggressive surgical treatment.

In the cohort study of Kaandorp et al [7], which comprised more than 3000 patients with joint pain evaluated in a rheumatology clinic, 37 patients were noted to have developed septic arthritis. They found that age exceeding 80 years, malignancy, diabetes mellitus, rheumatoid arthri- tis, use of immunosuppressive medication and history of joint replacement surgery were statistically significant risk factors for the development of septic arthritis [8]. In our case series, 84% of the documented cases of septic arthritis at our institution had 1 or more of these predisposing risk factors. Multiple-case series have documented other associated risk factors for septic arthritis, such as hemodialysis, human immunodeficiency virus/acquired immunodeficiency syn- drome, chronic, and cirrhosis [9]. In the study of Li et al [3], 36% of all patients with septic arthritis had either diabetes mellitus (22%) or human immunodeficiency virus/Acquired immunodeficiency syndrome (12%).

The most common organism causing septic arthritis is Staphylococcus species, which accounts for 60% to more than 80% of all cases of septic arthritis. In the 1997 study of Kaandorp et al [7] on 157 patients with septic arthritis,

S. aureus was the causative organism in 44% of cases. In their study, Li et al [3] observed that in 42 (76%) of 55 patients S. aureus was the causative organism [7,9,10]. Other common causative organisms are coagulase-negative staphylococcus, streptococcus, and gonococcus. Coagulase- negative staphylococcus is common in patients with recent joint replacement surgery. Dubost et al [9] has shown that incidence of coagulase-negative staphylococcus has in- creased from 11% in the 1980s to 21% of patients in the 1990s. They relate this to the increased prevalence of joint replacement surgery [9]. Methacillin-resistant S. aureus is becoming a common cause of septic arthritis in chronically ill and frequently hospitalized patients, as well as intrave- nous drug-using (IVDU) patients [11]. In the series of Dubost et al [9], the rate of methacillin-resistant S. aureus remained stable at 15% of all septic arthritis incidents over 20 years.

The joints most commonly involved in septic arthritis vary by study. Li et al [3] and McCutchan and Fisher [4] showed that the knee accounted for 36% to 55% of their total septic joint population. However, in the study by Li et al [3] the shoulder was the next most common joint involved (14%), and in the study by McCutchan and Fisher [4] the wrist ranked second (21%). In our study, the most common joint involved was the knee (65%), followed by the hip (12%). Spinal axial fibrocartilaginous joints are often the site of septic arthritis in IVDU patients [10]. The acromial-clavicular joint is often noted as a rare joint of septic arthritis, commonly associated with IVDU [11].

Patients generally present with the complaint of a painful, swollen joint. They may also have systemic symptoms including fever and chills. There may be a history of antecedent trauma to the area, or of recent illness. They will usually complain of increased pain with move- ment of the joint, and will often have increased discomfort with axial loads on the joint capsule. Often they will resist even passive movement of the joint [10].

Making the diagnosis of septic arthritis requires arthro- centesis of the suspected joint, and culture of the joint fluid. More than 20% of cases will have negative cultures including the culture for gonococcal arthritis, which is more frequently negative compared to other organisms [9]. Multiple studies have shown synovial WBC, erythrocyte sedimentation rate , and Gram’s stain to have poor sensitivity or specificity for identifying septic arthritis. Li et al [3] and McCutchan and Fisher [4] showed that the sensitivity of an elevated ESR was 90% to 96%. They also showed that the sensitivity of an elevated peripheral WBC was 48% to 57% [3,4]. In the study of McCutchan and Fisher [4], 50% of patients with septic arthritis had a synovial WBC of less than 28000/mm³. Shmerling et al

[12] showed the sensitivity of synovial WBC to be 84% with a specificity of 84% in diagnosis inflammatory arthritis. In our study, a synovial WBC of less than 50000 cells/mm³ lacked the required sensitivity to rule out septic arthritis. A sensitivity of 61% is too low to reliably exclude septic arthritis. Our study is clearly in line with recent research showing that patients with low synovial WBC counts can have culture-proven septic arthritis.

Multiple studies in the pediatrics literature looking at the difference between septic arthritis and Transient synovitis have led to the development and validation of clinical prediction rules [13-15]. The studies examining ancillary testing in predicting septic arthritis in adults have been unsuccessful to date in creating sensitive predictors of septic arthritis [3,4,16-18]. The use of ESR, peripheral WBC, and comorbidities to predict septic arthritis in adult patients has not achieved the desired sensitivity to rule out septic arthritis without obtaining joint fluid.

The mainstays for treatment of septic arthritis include a combination of Intravenous antibiotics with or without surgical intervention. There is debate in the literature

regarding open surgical drainage vs repeated Needle aspiration treatment. The definitive therapy for septic arthritis is often consultant-dependent [19].

Common antibiotic regimens are used to cover the most likely organisms. If the Gram’s stain is positive, it allows the clinician to tailor the antibiotic choice. If the Gram’s stain is negative, the clinician is left to treat on an empiric basis, knowing that the most common organisms are staph, strep, and gonococcus. Often, several antibiotics are used, including third-generation cephalosporins, fluroquinolones, aminoglycosides, semisynthetic penicillins, and-if the suspicion for methacillin-resistant S. aureus is high- vancomycin [9,10].

Limitations

Our study has several limitations. First, our study is a retrospective case series and, therefore, by design there is no way to calculate the specificity of the synovial WBC count. However, the design does allow for examination of the sensitivity of current guidelines regarding the diagnosis of septic arthritis. A second limitation is the narrow definition of septic arthritis; we included only cases of culture-proven septic arthritis. This likely eliminated in patients who are already on antibiotics with negative synovial cultures and in patients with a clinical diagnosis of septic arthritis by examination, blood culture, or surgical findings.

Conclusion

Our study shows that the evaluation of synovial fluid using a 50 000 WBC/mm³ cutoff for septic arthritis lacks accuracy. A sensitivity of 61% (95% CI, 48%-75%) is far too low on which to base reliable Clinical decisions. It is apparent that a lower cutoff (even at the expense of specificity) may be necessary to properly rule out septic arthritis, especially when significant predisposing medical conditions exist. A large prospective study is needed to better define the predictors of septic arthritis and to determine a clinical prediction rule based on a combination of ancillary tests, medical history, and synovial fluid analysis to better predict which patients will ultimately have culture-proven septic arthritis.

Septic arthritis is a debilitating disease and remains a challenging diagnosis. In accordance with our findings, we recommend that the emergency physician maintain a high

index of suspicions in at-risk patients and not use a synovial fluid WBC of less than 50000 WBC/ mm³ to rule out septic arthritis.

References

1. Hasselbacher et al. Arthrocentesis synovial fluid and synovial biopsy. Primer on rheumatologic disease. 11th edition, 1999.
2. Yu KH, et al. Concomitant septic and gouty arthritis-an analysis of 30 cases. Rheumatology (Oxford) 2003;42(9):1062 - 6.
3. Li SF, et al. Laboratory tests in adults with monoarticular arthritis: can they rule out a septic joint? Acad Emerg Med 2004;11(3): 276 - 80.
4. McCutchan HJ, Fisher RC. Synovial leukocytosis in infectious arthritis. Clin Orthop Relat Res 1990;(257):226 - 30.
5. Baer PA, et al. Coexistent septic and crystal arthritis. Report of four cases and literature review. J Rheumatol 1986;13(3):604 - 7.
6. Pioro MH, Mandell BF. Septic arthritis. Rheum Dis Clin North Am 1997;23(2):239 - 58.
7. Kaandorp CJ, et al. Incidence and sources of native and prosthetic joint infection: a community based prospective survey. Ann Rheum Dis 1997;56(8):470 - 5.
8. Kaandorp CJ, et al. Risk factors for septic arthritis in patients with joint disease. A prospective study. Arthritis Rheum 1995;38(12): 1819 - 25.
9. Dubost JJ, Soubrier M, Sauvezie B. Pyogenic arthritis in adults. Joint Bone Spine 2000;67(1):11 - 21.
10. Till SH, Snaith ML. Assessment, investigation, and management of acute monoarthritis. J Accid Emerg Med 1999;16(5):355 - 61.
11. Ross JJ, Shamsuddin H. Sternoclavicular septic arthritis: review of 180 cases. Medicine (Baltimore) 2004;83(3):139 - 48.
12. Shmerling RH, et al. Synovial fluid tests. What should be ordered? JAMA 1990;264(8):1009 - 14.
13. Del Beccaro MA, et al. Septic arthritis versus transient synovitis of the hip: the value of screening laboratory tests. Ann Emerg Med 1992; 21(12):1418 - 22.
14. Kocher MS, et al. Validation of a clinical prediction rule for the differentiation between septic arthritis and transient synovitis of the hip in children. J Bone Joint Surg Am 2004;86-A(8):1629 - 35.
15. Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence- based clinical prediction algorithm. J Bone Joint Surg Am 1999; 81(12):1662 - 70.
16. Schlapbach P, et al. Bacterial arthritis: are fever, rigors, leucocytosis and blood cultures of diagnostic value? Clin Rheumatol 1990;9(1): 69 - 72.
17. Soderquist B, et al. Bacterial or crystal-associated arthritis? Discrim- inating ability of serum inflammatory markers. Scand J Infect Dis 1998;30(6):591 - 6.
18. Gupta MN, Sturrock RD, Field M. A prospective 2-year study of 75 patients with adult-onset septic arthritis. Rheumatology (Oxford) 2001;40(1):24 - 30.
19. Goldenberg DL, et al. Treatment of septic arthritis: comparison of needle aspiration and surgery as initial modes of joint drainage. Arthritis Rheum 1975;18(1):83 - 90.