In our study, the LE test had acceptable sensitivity and specificity for SA detection. When (+) and (++) were both considered positive, the sensitivity and specificity and accuracy of the LE test for the detection of SA were 100% and 43.33% (95% CI: 25.6–61.0%) and 63.04% (95% CI: 48.9–77.2%), respectively. When only (++) was considered positive, the sensitivity, specificity, and accuracy of the LE test for the detection of SA were 93.8% (95% CI: 81.8–100%), 60% (95% CI: 42.4–77.5%), and 71.7% (95% CI: 58.5–84.9%), respectively. Our results show that with an AUC of 0.77, LE test can be a useful diagnostic method when used along with SF culture to provide a fast and sensitive diagnostic method for clinical decision making in the time it takes for culture results.

Diagnosing SA can be challenging since various differential diagnoses can manifest as SA. Patients typically do not present with the typical signs of infection. These vague clinical manifestations can mislead physicians to other differential diagnoses, such as Perthes’s disease or juvenile idiopathic arthritis (JIA) [16]. The current clinical and laboratory tests for SA are either time-consuming or do not have high sensitivity and specificity. In 1976, Newman proposed a set of diagnostic criteria to diagnose bacterial arthritis, including whether any organism was isolated from the joint, from blood, or histological or radiological findings of infection even in the case of no organism isolated or turbid fluid aspiration [24]. Currently, the most reliable methods to diagnose SA are SF aspiration studies, such as smear and culture. Various laboratory markers aid clinicians in diagnosis and can be used in conjunction with SF culture studies.

ESR and CRP are two of the most commonly used laboratory tests for SA diagnosis. Kocher’s criteria, being the most accepted diagnostic criteria for diagnosing SA in children, use WBC count > 12,000 cells/mm3, ESR > 40 mm/hour, fever > 38.5 °C, and whether the limb is weight bearing or not. In case one criterion exists, there is a 3% chance of SA; in the presence of two, 40%; in the presence of three criteria, 93%; and in case all four criteria are positive, there is a 99% chance of SA [25]. The sensitivity and specificity of serum ESR have been estimated to be between 34% and 100% and 23% and 93%, respectively [9, 26,27,28,29]. Serum CRP has been widely studied as a diagnostic marker for SA. Its sensitivity and specificity vary in different studies depending on the cut point between 58% and 100% and 0% and 96%, respectively [9, 12, 26, 28,29,30,31,32]. Although used in Kocher’s criteria, the serum WBC count is not a very sensitive marker since its sensitivity and specificity range between 20 and 62% and 61–100%, respectively [9, 12, 27,28,29,30,31,32]. SF WBC counts with a cutoff point of 50,000 cells/mm3 have a sensitivity of 54–100% and a specificity of 66–97% [9, 27, 28, 30, 31]. Procalcitonin (PCT), a relatively novel marker, can be measured in serum and SF. In SF, it has a sensitivity of 17–87% and a specificity of 55–100% [9, 27,28,29,30]. In serum, it has shown a sensitivity of 80–100% and specificity of 68–100% [9, 28, 29, 33]. There are many other markers, such as SF PMN%, glucose, LDH, calprotectin, IL-6, uric acid, lactate, serum IL-6, TNF-alpha, uric acid, and many others, which can be utilized together with other diagnostic tools [9].

Since SA is a debilitating disease that needs timely management, a late or underdiagnosis is unacceptable in the case of SA. Thus, a sensitive and accurate diagnostic test is paramount. This study investigated the LE strip test to potentially improve the orthopedic surgeon’s diagnostic ability regarding timeliness and accuracy. The LE dipstick test is relatively fast and affordable for detecting leukocytes in body fluids. By detecting the presence of the esterase enzyme produced by leukocytes, it was previously used for detecting UTIs and other infections.

Some studies have previously investigated the role of the LE dipstick test in diagnosing septic arthritis (Table 3). Knapper et al. investigated the diagnostic accuracy of the LE test in eighty adult patients (74% men, 26% women) in three hospital sites between 2015 and 2016. LE results were reported as negative (n = 9, 11%), + (n = 14, 18%), ++ (n = 24, 30%), and +++ (n = 33, 41%). Negative and trace (+) results were considered negative, and 57 patients had positive (++, +++) LE results. Five patients had positive 48-hour cultures. Of the remaining 52, 34 were diagnosed with crystal arthropathy, 17 were presumed to have an arthritic flare, and no cause was identified in one case. Their study yielded the results of 100% sensitivity, 30.7% specificity, 8.77% PPV, and 100% NPV [34]. A summary of the diagnostic accuracy of LE in SA diagnosis can be seen in Table 3. LE’s sensitivity (ranging from 80 to 100%) and specificity (ranging from 30 to 100%) in diagnosing SA were high, even more than in some other tests. Therefore, due to its high sensitivity, it can be used as a valuable screening tool. As it can be seen in Table 3, in the studies conducted by Colvin et al. and Knapper et al., LE test has a sensitivity of 100% and in the study by Omar et al. and Kolbeck et al. the sensitivity of LE test is around 95% which combined with the high specificity of in the seven out of the eight similar studies, shows that LE test can be used in conjunction with the other diagnostic methods to facilitate diagnosis and treatment, the differences between the diagnostic accuracy measures between studies can originate from different factors such as variability in LE reagent strips, their interpretation, the time needed before reading the strip, the prevalence of the condition in the population, other confounding factors and other diagnosis that can lead to the presence of WBC in the synovial fluid and thus yielding a positive result for the LE test, such as inflammatory arthritis diseases.

The LE test can be performed and returns a reliable result within a minute or two. This can also be useful when the clinician faces doubt regarding SA preoperatively. Parvizi et al. previously investigated the accuracy of the LE test in detecting the cause of periprosthetic joint infection (PJI) [19]. In the study conducted by Parvizi et al., a (++) reading on the LE test meant that the diagnosis of PJI was nearly confirmed, and a negative or trace result almost ruled out septic arthritis [19]. Based on the results of our study and other researchers’ studies, we can claim that the LE dipstick, either by itself or in conjunction with other diagnostic methods, can be used as a screening test for SA in suspected patients. Since LE is a product of leukocytes, it can be said to be a proxy for detecting WBCs in the SF. As Parvizi et al. claimed, detecting LE instead of WBC and measuring PMN percentage has a fascinating advantage in that it can include dead cells. In contrast, WBC measurement does not include dead cells in the case of very aggressive infections, and there are a high number of dead WBCs in the SF [19]. Therefore, the LE test can be used initially to screen SA before a more time-consuming and expensive workup is needed.

The LE test has some negative points. One limitation of the LE dipstick test is that it requires the collection of synovial fluid, which can be painful for patients since it is an invasive procedure. Additionally, the LE test is subject to false positives and negatives, which can lead to unnecessary treatment or delayed diagnosis. False positives can occur in patients with other inflammatory joint diseases, such as rheumatoid arthritis, gout, or reactive arthritis; in contrast, false negatives can occur in patients with septic arthritis who do not have leukocytosis or those who receive antibiotic treatment before the test [35].

Our study faced some limitations. First, the qualitative nature of LE test results might make it challenging to suggest a precise cutoff value of LE enzyme in SF to rule out or confirm SA diagnosis. Second, there are various protocols for administering LE, varying the time the dipstick is in the sample and the number of tries. Consequently, our protocol may differ from previous protocols, indicating heterogeneity between the studies. Conducting studies with larger sample sizes can help determine the exact diagnostic accuracy of the LE test. Also, since we could not indicate an aggressive procedure like joint fluid aspiration for healthy people our controls were patients undergoing knee arthroplasty and this makes our study susceptible to Berkson bias. The primary goal of our study was to assess the diagnostic accuracy of LE dipstick in early diagnosis of SA and we did not further gather for other diseases such as RA, AS, gout, etc. and excluded patients with any probable diagnosis other than SA from our study which would be valuable for future studies to inquire about.