Benign intra-articular soft tissue tumors mimicking hip septic arthritis: an arthroscopic case series with tenosynovial giant cell tumor predominance

Abstract

Background

Benign intra-articular soft tissue tumors can mimic septic arthritis of the hip (SAH), posing a significant diagnostic challenge. Timely differentiation between these entities is crucial, as SAH requires urgent intervention to prevent adverse outcomes. The present study aims to (1) examine clinical presentations, imaging findings, laboratory results, and histopathological features associated with benign intra-articular soft tissue tumors mimicking SAH and (2) describe diagnostic challenges encountered in treating these patients.

Methods

A retrospective review identified 23 patients who underwent arthroscopic irrigation and debridement for suspected SAH at a single institution between 2014–2023. Seven patients with negative cultures and intraoperative diagnoses of intra-articular tumors confirmed by histopathology were included in the primary analysis. Clinical presentations, imaging findings, laboratory results, arthroscopic findings, histopathology results, and follow-up data were analyzed. A comparison with the non-tumor cases (n = 16) was performed as a secondary analysis.

Results

The mean age of the tumor group was 25.0 ± 8.1 years and a female predominance (6/7, 85.7%) was observed. All patients presented with acute hip pain, limited range of motion, and elevated C-reactive protein levels (80.0 ± 38.3 mg/L). Three patients presented with fever and three with leukocytosis. Synovial fluid white blood cell counts exceeded 50,000 cells/μL in five patients. Arthroscopy revealed solid tumors in all cases. Tumor necrosis was observed in all cases and tenosynovial giant cell tumor (TGCT) was confirmed histologically in six cases (6/7, 85.7%). Postoperative MRI in four patients showed no tumor recurrence. Compared to the non-tumor group, patients with tumors were significantly younger (p = .016) and had lower preoperative CRP levels (p = .027).

Conclusions

Differentiating benign intra-articular tumors from SAH remains difficult, particularly among younger and female patient populations with acute presentations of TGCT which, as observed in the present study, may be explained by histopathologic evidence of tumor pedicle torsion and resultant necrosis. MRI can be valuable in identifying intra-articular lesions given that clinical and laboratory findings are frequently equivocal. The proposed diagnostic algorithm offers a framework for identifying cases where preoperative MRI may be beneficial, though it requires validation in larger studies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12891-026-09892-w.

Background

Septic arthritis of the hip (SAH) is a critical surgical emergency that, without prompt intervention, can lead to rapid joint deterioration, septicemia, and potentially fatal outcomes [23, 26]. SAH is characterized by the acute onset of hip pain, stiffness, limp, and fever [7, 16]. Various non-infectious conditions can mimic the clinical presentation of SAH, posing diagnostic challenges and potentially leading to misdiagnosis and inappropriate treatment [8, 29].

Formerly known as pigmented villonodular synovitis (PVNS) [4, 35], tenosynovial giant cell tumor (TGCT) is characterized by the proliferation of synovial-like mononuclear cells, multinucleated giant cells, and hemosiderin-laden macrophages [25, 27, 35]. TGCT is a rare, benign proliferative disorder of the synovial membrane that typically presents with chronic joint pain and swelling [11]. Recent case reports, particularly in the pediatric population, have described acute presentations of TGCT mimicking SAH due to overlapping features such as pain, fever, stiffness, limited range of motion, elevated inflammatory markers, and leukocytosis [10, 14, 24, 34, 39]. However, such acute presentations of TGCT and other benign intra-articular soft tissue tumors in the hip appear to be rare in adults, and published data remains limited [10, 14, 24, 34].

The present case series aims to examine the clinical presentations, imaging findings, laboratory results, and histopathological features of benign intra-articular soft tissue tumors mimicking SAH. The suspected pathophysiological mechanism underlying the acute presentation of these tumors and our treatment experience are discussed.

Methods

Study design and patient selection

A retrospective review of patients who underwent arthroscopic irrigation and debridement for suspected septic arthritis of the hip (SAH) was conducted at a single tertiary medical center. Institutional Review Board approval was obtained prior to study commencement (IRB approval: TLV-0402–13, Clinical Trial Number: not applicable) and all patients consented to participation. The decision to perform arthroscopic surgery was based on the institution's standard practice for managing suspected SAH, which includes clinical presentation, laboratory findings, and imaging results suggestive of infection. Patients were not excluded on the basis of medical comorbidities or pregnancy. A total of 23 cases of suspected SAH that underwent arthroscopic hip irrigation and debridement were identified from two fellowship-trained senior hip arthroscopists’ practices from January 2014 to April 2023. Of these, seven patients with intra-articular tumors diagnosed intraoperatively and confirmed by histopathology were included in the primary analysis. The remaining 16 patients without intra-articular tumors formed a secondary comparative cohort.

Data collection

Data were retrospectively collected from the institution’s electronic medical records. Data collection included patient demographics, clinical presentation, laboratory and microbiological assays (blood tests and synovial fluid analyses), imaging studies, arthroscopic findings, and histopathological results.

Diagnostic criteria

Based on criteria proposed in prior studies, suspected SAH was defined as having (1) at least two of the following clinical manifestations: hip pain, limited range of motion (ROM), inability to bear weight, and fever (> 38 °C); and (2) at least one of the following laboratory findings: elevated serum white blood cell (WBC) count over 11,000 cells/µL, elevated serum C-reactive protein (CRP) level over 5 mg/L, abnormal synovial fluid analysis, or a positive synovial fluid Gram stain [6, 7, 20, 22, 26]. All patients who met diagnostic criteria for SAH were indicated for irrigation and debridement within 48 h of presentation.

Demographics and clinical data

Demographic data included sex, age, and relevant medical history. Preoperative clinical data included reported hip pain, initial ROM, ability to bear weight, fever (temperature ≥ 38.0 °C) at presentation to the emergency department (ED), and the duration of symptoms before ED admission. Postoperative joint function was evaluated using the International Hip Outcomes Tool (iHOT-33) at the last follow-up visit. Preoperative (baseline) iHOT-33 was deferred, as acute SAH involves symptom fluctuations that render a time-point analysis of clinical function challenging and not representative of the patient’s clinical state. Additional data extracted from follow-up visits included the duration of post-surgery antibiotic treatment and time to recovery of full ROM.

Laboratory and microbiological assays

Blood tests

Preoperative blood tests (part of the ED and inpatient orthopaedic workup) included complete blood count (CBC) with differential for WBC count, CRP levels (reference range: 0–5 mg/L), and blood cultures. Postoperative blood tests included CBC with differential for WBC count until discharge, CRP levels until discharge, and both WBC counts and CRP levels weekly until completion of antibiotic treatment. CRP levels and WBC count at discharge or first follow-up (two weeks post-surgery) were used to assess treatment effectiveness.

Synovial fluid analysis

Synovial fluid analyses were conducted preoperatively with ultrasound-guided aspiration in the ED and intraoperatively with direct fluid visualization in the operating room. Fluid analyses included WBC count (cells/μL), direct microscopic smear (Gram stain), and microbiological cultures with broad-spectrum or "pan-microbial" polymerase chain reaction (PM-PCR) [18] performed only at the request of the infectious disease specialist. Synovial fluid samples were classified as "normal," "inflammatory," "septic," or "hemorrhagic" based on the color, white blood cell count, and gram stain/culture as used by Honig et al. [14, 33, 37]

Imaging studies

Standardized imaging obtained in the ED included routine pelvic radiographs and hip ultrasonography for the detection of joint effusion with subsequent ultrasound-guided aspiration if an effusion was identified. Additional imaging, if performed, including computed tomography (CT) examinations and magnetic resonance imaging (MRI) examinations, was reviewed; however, initial advanced imaging was not routinely obtained given the current guidelines in support of obtaining arthrocentesis rather than advanced imaging in the diagnosis of SAH [2, 9, 23, 30]. Postoperative MRI was recommended for all patients in the tumor group and completed in four patients.

Surgical technique

All surgeries were performed by two fellowship-trained hip arthroscopists with over ten years of experience. Patients were positioned supine with the hip flexed to 45 degrees. A proximal anterolateral portal (PAP [38]) was first established for visualization of the peripheral compartment, followed by a distal anterolateral portal under arthroscopic visualization. A longitudinal interportal capsulotomy was performed, followed by diagnostic arthroscopy of the peripheral compartment to evaluate the anteromedial neck, posteromedial gutter, femoral head and labrum, anterolateral neck, posterolateral gutter, distal neck, and capsule. Intra-articular debris was systematically removed using a 3.5 mm full-radius resector shaver (Dyonics; Smith & Nephew, Andover, MA, USA) from the capsular side, head, and neck, starting at the anteromedial joint gutter and proceeding to the posteromedial, anterior, anterolateral, posterolateral, and anterior neck and capsule. In all cases among patients with intraarticular tumors, a well-circumscribed, pedunculated soft tissue mass was observed emanating from the capsular side. When present, the mass was excised at its pedicle base using a hooked radiofrequency probe (VAPR; DePuy Mitek, Raynham, MA, USA) and submitted for gross pathological and histological examination. The joint was irrigated with 9–12 L of saline. An intraarticular drain was placed and removed after 24–48 h.

Arthroscopic, macroscopic, and histopathological findings

Intraoperative data included arthroscopic surgical videos and photographs as well as all operative reports. Macroscopic and histopathological findings were obtained from pathology reports of the excised specimens. The pathology specimens were reviewed again for this study by a senior pathologist specialized in musculoskeletal pathology.

Statistical analysis

Descriptive statistics were reported as mean ± standard deviation or median and range for continuous variables and as frequencies and percentages for categorical variables. The choice between mean and median was made based on the distribution of the data, with medians used for non-normally distributed variables. For the secondary comparative analysis, the normality of data distribution was assessed using the Shapiro–Wilk test. Due to the small sample size and potential non-normal distribution of data, non-parametric tests were used for comparing continuous variables between groups, except for age where both parametric and non-parametric tests were performed. For age comparison, independent t-tests were used if normality was confirmed; otherwise, the Mann–Whitney U test was employed. The Mann–Whitney U test was used for comparing C-reactive protein (CRP) levels, blood white blood cell (WBC) count, and synovial fluid WBC count between groups. Fisher's exact test was used for categorical variables due to the small sample size.

A receiver operating characteristic (ROC) curve analysis was performed to determine the optimal C-reactive protein (CRP) threshold for discriminating between tumor and non-tumor cases. The area under the curve (AUC) was calculated to assess the discriminatory ability of CRP. The performance of the proposed diagnostic algorithm was evaluated by calculating its sensitivity and specificity when applied to our study population. All statistical analyses were performed by the authors using IBM SPSS Statistics (Version 26, IBM Corp., Armonk, NY, USA). A two-tailed p-value ≤ 0.05 was considered statistically significant.

Results

Demographics

Of the 23 patients presenting with symptoms concerning SAH, seven were found to have masses present on arthroscopic evaluation and were included in the tumor group. The mean age in this group was 25 ± 8.1 years and six patients (85.7%) were female (Table 1). All patients were otherwise healthy with one noted to be 39 weeks pregnant. The mean follow-up time was 22 ± 7.0 months.

Table 1.

Patients with intra-articular tumors: patient demographics and preoperative characteristics

	Sex	Age (Years)	Primary Symptoms	History	Days of Symp prior ED	Fever* at ED °C	Peripheral WBC	CRP (mg/L)	Synovial WBC (Gross appearance)	MRI (Diagnosis)	Synovial Gram & culture
Pt. 1	F	20	Pain; ROM	Atraumatic	0	(38.4 °C)	15.2	49	50,500 (Septic)	No	Neg
Pt. 2	F	34	Pain; ROM	Atraumatic	1	Afebrile	14.0	65	82,500 (Septic)	No	Neg
Pt. 3	F	33	Pain; ROM	Atraumatic; 29 weeks pregnant	1	Afebrile	10.7	124	120,000 (Septic)	No	Neg
Pt. 4	F	28	Pain; ROM	E-scooter fall 2 wks prior to onset of pain	7	Afebrile	8.2	35	38,000 (Inflam.)	Yes (mass)	Neg
Pt. 5	F	14	Pain; ROM	Atraumatic	2	(38.1 °C)	11.6	120	42,000 (Septic)	Yes (mass)	Neg
Pt. 6	F	29	Pain; ROM	Atraumatic	4	Afebrile	8.1	112	115,000 (Septic)	Yes (mass)	Neg
Pt. 7	M	17	Pain; ROM	Atraumatic; 3 mo of chronic right groin pain	3	(38.2 °C)	7.7	55	55,500 (Inflam.)	Yes (mass)	Neg

Pt Patient, ED Emergency Department, Symp. Symptoms, ROM Range of Motion, WBC White Blood Cell count (× 10³ cells/μL), CRP C-reactive Protein (mg/L), Synovial WBC Synovial fluid White Blood Cell count (cells/mL), MRI Magnetic Resonance Imaging, Neg Negative, Wks Weeks, Inflam Inflammatory

^*Fever defined as body temperature ≥ 38.0 °C

The remaining 16 patients, confirmed to have SAH without intra-articular tumors, formed the non-tumor group. This group had a significantly higher mean age of 43.3 ± 17.2 years and a male predominance (n = 10/16, 62.5%) (Table 2).

Table 2.

Comparison of intra-articular tumor and non-tumor patient groups

Characteristic	Tumor Group (n = 7)	Non-Tumor Group (n = 16)	P-value
Age (years), mean ± SD	25.0 ± 8.1	43.3 ± 17.2	0.016*
Gender, n (%)			0.094
- Female	6 (85.7%)	6 (37.5%)
- Male	1 (14.3%)	10 (62.5%)
Blood WBC count (× 103 cells/μL), mean ± SD	10.8 ± 3.1	11.4 ± 4.2	0.742
Preoperative CRP (mg/L), mean ± SD	80.0 ± 38.3	171.0 ± 132.8	0.027*
Synovial fluid WBC count (cells/mL), mean ± SD	71,929 ± 36,325	98,333 ± 77,606	0.526
Fever present, n (%)	3 (42.9%)	9 (56.3%)	0.667

WBC White Blood Cell, CRP C-reactive Protein, SD Standard Deviation

^*Statistically significant (p < 0.05)

Clinical data

All patients presented with acute, non-traumatic hip pain and significant limitations in hip ROM. Three patients presented with fever (temperature ≥ 38.0 °C), with temperatures ranging from 38.1 °C to 38.4 °C. The duration of symptoms before ED presentation ranged from zero to seven days, with a median of two days (Table 1). All seven patients showed clinical improvement after arthroscopic intervention, with resolution of fever (n = 7/7), and minimal to no limitations in ROM by discharge (n = 4/7; 2–8 days postoperatively) or by the first follow-up visit at two weeks postoperatively (n = 3) (Table 3).

Table 3.

Patients with intra-articular tumors: intraoperative findings and postoperative characteristics

	Arthroscopy Findings	Operative Culture	Pathological report	Discharge or 1 st FU CRP (mg/L)	Total Abx Tx (Weeks)	MRI post-op	FU time (Months)	iHot-33 score last FU
Pt. 1	Villonodular mass	Negative including PM-PCR	TGCT; ischemic necrosis	< 5 (1st FU)	1.5	No	17.7	100
Pt. 2	Solid nodular mass	Negative	Suggestive of TGCT; ischemic necrosis; neutrophils	< 5 (POD 8)	2.5	Yes	17.6	92
Pt. 3	solid mass	Negative	Suggestive of TGCT; ischemic necrosis; neutrophils	< 5 (1st FU)	0.5	No	26.9	100
Pt. 4	Villonodular mass with a torsed pedicle	Negative including PM-PCR	Almost complete ischemic necrosis; neutrophils	< 5 (1st FU)	2	Yes	16.8	87
Pt. 5	synovial solid mass	Negative including PM-PCR	Suggestive of TGCT; ischemic necrosis; neutrophils	< 5 (POD 6)	3	Yes	28.8	100
Pt. 6	Villonodular mass	Negative including PM-PCR	TGCT; ischemic necrosis; neutrophils	5 (POD 5)	0.5	No	32.1	93
Pt. 7	Nodular diffuse masses	Negative	TGCT; ischemic necrosis	< 5 (POD 5)	1	Yes	14.3	95

Pt Patient, WBC White Blood Cell, CRP C-reactive Protein, POD Postoperative Day, PM-PCR Pan-Microbial Polymerase Chain Reaction, LOS Length of Stay, OP Surgical, FU Follow-up, Abx Antibiotics, Tx Treatment, TGCT Tenosynovial Giant Cell Tumor, MRI Magnetic Resonance Imaging, Neg Negative, iHOT-33 International Hip Outcome Tool

Laboratory and microbiological assays

All patients presented with elevated blood CRP levels (> 5 mg/L; reference range: 0–5 mg/L), with a median level of 65 mg/L (range: 35–124 mg/L). Three patients (42.9%) demonstrated leukocytosis (WBC count > 11,000 cells/μL; reference range: 4,000–11,000 cells/μL) at time of admission (Table 1). Marked effusion on ultrasonographic evaluation was evident in all patients. Synovial fluid aspirate (Fig. 1) demonstrated variability in clarity and gross appearance, although most cases (n = 5/7, 71.5%) could be classified as having a "septic" appearance (i.e., opaque yellowish appearance) according to Honig et al. [14] Synovial fluid WBC counts exhibited considerable variability, ranging from 38,000 to 120,000 cells/μL (median: 55,500 cells/μL), exceeding the 50,000 cells/μL threshold typically suggestive of SAH [7, 14, 20, 22] in five patients. After arthroscopic intervention, substantial decreases in WBC count and CRP levels were observed among all patients. All patients were discharged without leukocytosis. CRP levels returned to normal ranges prior to discharge in 57% (n = 4/7) of patients or by two-week postoperative follow-up in 43% (n = 3/7). All preoperative and intra-operative synovial fluid Gram stains and cultures were negative.

Fig. 1.

Synovial fluid aspirated, gross appearance. A opaque yellowish (Patient #1); B opaque yellowish-bloody (Patient #2); C translucent yellowish (Patient #4)

Table 4 presents a summary of the associated clinical and laboratory findings.

Table 4.

Patients with intra-articular tumors: overlapping septic arthritis symptoms

		Pt. 1	Pt. 2	Pt. 3	Pt. 4	Pt. 5	Pt. 6	Pt. 7	Total pt. (n/7)
Clinical	Limited hip ROM	+	+	+	+	+	+	+	7/7 (100%)
	Fever	+				+		+	3/7 (43%)
Blood	Leukocytosis*	+	+			+			3/7 (43%)
	Abnormal CRP	+	+	+	+	+	+	+	7/7 (100%)
Synovial	Septic Gross appearance	+	+	+		+	+		5/7 (71.5%)
	WBC > 50,000 (cells/μL)	+	+	+			+	+	5/7 (71.5%)
Total signs per patient (n/6)		6/6	5/6	4/6	2/6	5/6	4/6	4/6

SAH septic arthritis of the hip, Pt Patient, ROM Range of Motion, CRP C-reactive Protein, WBC White Blood Cell

^*Leukocytosis defined as WBC count > 11,000 cells/μL

Imaging

Routine pelvic radiographs of all patients in the ED were unremarkable. All patients underwent hip ultrasonography for the detection of joint effusion, followed by arthrocentesis. Marked effusion was evident in all sonographic examinations.

Five patients underwent contrast-enhanced CT that demonstrated nonspecific joint effusion and synovial enhancement. Four patients underwent hip MRI during hospitalization, revealing intra-articular or intra-capsular lesions (Fig. 2A).

Fig. 2.

Gross and microscopic mass appearance. A Proton Density-Weighted Coronal MRI Image: demonstrates a 1.3 × 2.2 × 3.6 cm intraarticular mass (arrows) in the inferior-medial aspect of the joint with heterogeneous signal intensities. B Arthroscopic Appearance of Tumor: demonstrates a nodular tumor with torsion of the pedicle (arrows). C Gross Appearance of Excised Tumor. D Microscopic Histopathologic Appearance (Hematoxylin and Eosin, × 10): demonstrates proliferation of mononuclear cells and giant cells (black arrows) set in collagenous stroma with hemosiderin pigment (white arrows) consistent with a tenosynovial giant cell tumor (TGCT). E Microscopic Histopathologic Appearance (Hematoxylin and Eosin, × 20): demonstrates ischemic necrosis and congested blood vessels (arrows) due to torsion. F Microscopic Histopathologic Appearance (Hematoxylin and Eosin, × 20): demonstrates a collection of neutrophils observed in the tumor tissue

Arthroscopic, macroscopic, and microscopic findings

Arthroscopy revealed solid tumors at the joint periphery in all patients, commonly with the presence of villonodular masses, tumor pedicle torsion, and ischemic necrosis (Table 3, Fig. 2B-C, Additional Video 1). Macroscopically, clear torsion of the tumor pedicle was evident on gross examination in one patient. Microscopically, most lesions (n = 6/7, 85.7%) exhibited histological features characteristic of TGCT including mononuclear cells, giant cells, and hemosiderin pigment; however, one sample (#4) demonstrated near-complete ischemic necrosis of the excised lesion that impeded further characterization. All excised tumors demonstrated scattered areas of ischemic necrosis to varying degrees, and 5 of the 7 tumors demonstrated areas with localized neutrophil collections (Fig. 2D-F).

Follow-up

At the 1-month, 6-month, and 18- 24-month follow-up visits, patients demonstrated no evidence of recurrent symptomatic pathology, reporting an average iHOT-33 score of 95 at final follow-up (Table 3).

During the follow-up period, all microbiological cultures and broad-spectrum PCR tests yielded negative results for infectious agents.

Four patients underwent postoperative hip MRI examinations between three and five months postoperatively, which showed small hematomas or minimal effusion with no evidence of tumor recurrence (Table 3).

Group comparison

Table 2 provides a detailed comparison of demographic and laboratory findings between the tumor and non-tumor groups. Patients in the tumor group (n = 7) were significantly younger than those in the non-tumor group (n = 16) (25.0 ± 8.1 vs 43.3 ± 17.2 years, p = 0.016). The tumor group also had significantly lower preoperative CRP levels compared to the non-tumor group (80.0 ± 38.3 vs 171.0 ± 132.8 mg/L, p = 0.027). There was a trend towards more females in the tumor group, but this did not reach statistical significance (85.7% vs 37.5%, p = 0.094).

No significant differences were found in serum WBC count (10.8 ± 3.1 vs 11.4 ± 4.2 × 10³ cells/μL, p = 0.742) or synovial WBC count (72,000 ± 36,000 vs 98,000 ± 78,000 cells/mL, p = 0.526) between the groups. The presence of fever was also not significantly different between the groups (42.9% vs 56.3%, p = 0.667). The duration of symptoms prior to ED presentation did not differ significantly (p = 0.410) between the tumor (2.57 ± 2.44 days) and non-tumor groups (3.80 ± 3.51 days).

Discussion

Benign intra-articular tumors, particularly tenosynovial giant cell tumor (TGCT), can present acutely with clinical and laboratory features mimicking septic arthritis of the hip (SAH). Among the 30% (n = 7/23) of patients who underwent hip arthroscopy for clinically suspected SAH and were found to have benign intraarticular tumors, over half (n = 4/7, 57.1%) presented with fever and/or peripheral leukocytosis, all presented with elevated CRP, and five demonstrated synovial fluid WBC counts above the threshold classically considered diagnostic of infection. The acute nature of symptom onset within one week of presentation coupled with the histopathologic findings of ischemic necrosis in all patients and gross torsional deformity clearly observed in one may explain the clinical symptomology, with tissue necrosis and torsional deformation of the tumor pedicle leading to an acute, intraarticular inflammatory response yielding symptoms similar to septic arthritis.

Across existing literature, there is considerable debate regarding adjunctive diagnostic methods and consistent surgical indications in the treatment of SAH [7, 12, 16, 20, 22, 26]. At this time, the gold standard for diagnosing septic arthritis (SA) is a positive synovial fluid culture; however, the delay of up to 7 days [21, 23, 26, 31] required to obtain positive cultures creates a disconnect in obtaining an accurate diagnosis of SAH and facilitating prompt surgical intervention for this medical emergency requiring immediate treatment. Subsequently, the decision to proceed with surgical intervention is often based on clinical, laboratory, and imaging parameters that provide a high suspicion of SAH, even prior to a definitive diagnosis.

However, basing a differential diagnosis of SAH on such clinical factors in an urgent setting can sometimes lead to the misdiagnosis of other conditions presenting with similar symptoms, such as benign tumors. Benign soft tissue tumors, such as TGCT, typically present with chronic joint pain and swelling, with an average duration of symptoms of 10–15 months before diagnosis [11, 17, 28, 36]. The knee is the most affected joint, accounting for two-thirds of cases, followed by the ankle and hip [35]. Acute presentation mimicking infection is exceptionally rare in TGCT and, consequently, not routinely considered in a differential diagnosis during the workup of SAH. Several case reports have described TGCT in the pediatric population presenting with acute hip pain closely mimicking SAH [14, 24, 32, 34], but there have been no similar reported findings among adults.

In the present study, six out of seven patients exhibited at least four clinical and laboratory findings suggestive of SAH, and five had synovial fluid WBC counts exceeding 50,000 cells/μL. Classically, synovial WBC counts over a 50,000 threshold are considered one of the best diagnostic tools for detecting septic arthritis, alongside positive synovial fluid gram stain and cultures [12, 22, 23, 26]. Despite clinical presentations, laboratory results, and imaging findings concerning for SAH, the absence of positive cultures in all cases of the present study highlights the diagnostic challenge and the potential for misdiagnosis due to the overlap in presentations. Similarly, intraoperative arthroscopic findings of tumor pedicle torsion and subsequent histological tumor necrosis due to infarction (Fig. 2), a mechanism previously suggested but rarely observed [10, 13, 15], offer an explanation regarding the presentation similarities between SAH and TGCT. Upon secondary review by a senior pathologist specialized in musculoskeletal pathology, all excised tumors showed scattered areas of ischemic necrosis to varying degrees, with 5 out of 7 tumors demonstrating neutrophil collections. The confirmation of TGCT in 6 cases through a postoperative histopathological evaluation, along with observed areas of ischemic necrosis and neutrophil collections, suggests an inflammatory response leading to neutrophil chemotaxis that may explain the elevated synovial fluid WBC counts and aligns with a similar hypothesis proposed by Saadat et al. [32]. This understanding provides insight into the observed clinical manifestations that resemble SAH and subsequently obfuscate the diagnosis of intraarticular TGCT of the hip.

The implication of the finding that at least 30% of patients initially suspected to have joint infection had a benign tumor is twofold:

1. Diagnostic Challenge: Distinguishing between joint infection and benign tumors can be difficult due to overlapping symptoms and clinical presentation. Thorough diagnostic evaluation, including appropriate imaging studies and possibly biopsy, is essential for accurate diagnosis.

2. Clinical Management: Misdiagnosis can result in inappropriate treatment, leading to delays in managing benign tumors or unnecessary interventions for presumed infections. A misdiagnosis of SAH in the setting of a benign soft tissue tumor or pseudotumor is substantial, as patients may be subjected deficient tumor resection and unnecessary prolonged antibiotic treatment of 3 to 6 weeks with risk of associated adverse effects [3, 5, 26]. The large proportion (n = 7/23) of benign lesions in a cohort of patients suspected to have SAH suggests that this diagnosis may be more common than previously documented, although the rate of benign lesions resembling SAH is not reported in existing literature, which mostly consists of case reports [10, 11, 14, 19, 24, 34, 39]. Clinicians must maintain a high index of suspicion for benign tumors, especially when the clinical response to standard infection management is poor or atypical features are present.

Despite the low volume of patients in the present study that precluded the ability to reach statistically significant conclusions regarding patient demographics, a female predominance (n = 6/7) was observed, with a mean age of 25 (14–34) years. Classically, SAH considered to have a male predominance, with a systematic review of 1,236 patients published in 2021 describing a 55% male population and a higher mean age of 49.9 (24–65) years [1]. Unsurprisingly, the demographics of the present population align more closely with patients with TGCT, as described in the latest international consensus paper published in 2023 [35] to include a higher prevalence among females and a lower mean age of 35–50 years at diagnosis. Given the evidence for demographic differences between patients with SAH and TGCT, these findings may assist clinicians in their diagnostic considerations when evaluating patients with suspected SAH.

Regarding adjunctive imaging in the diagnosis of SAH versus TGCT, preoperative MRI may offer value in diagnostic accuracy. At the authors’ institution, MRI is not part of the routine workup prior to open irrigation and debridement for suspected septic arthritis, a phenomenon attributable to the acute nature of presentation and logistical challenges associated with obtaining an MRI in a timely manner. However, among the four patients who underwent preoperative MRI due to increased clinical suspicion for an atypical diagnosis, MRI findings played a pivotal role in identifying intra-articular lesions, thereby reinforcing its value in distinguishing benign tumors from SAH when clinical and laboratory findings are equivocal and routine radiographs, ultrasounds, and CT evaluations are nondiagnostic [8, 10, 14, 34]. Subsequently, these findings suggest that MRI should be considered an essential component of the diagnostic workup in cases where suspicion for SAH remains high but the diagnosis is uncertain to prevent the sequelae of misdiagnosis and reduce the urgency for surgical intervention to allow for more thorough surgical planning and patient education.

Clinical implications and proposed diagnostic algorithm

This study highlights the diagnostic challenge in differentiating benign intra-articular tumors, particularly TGCT, from SAH. To address this challenge, we propose a diagnostic algorithm (Fig. 3) that may aid clinicians in identifying cases where preoperative MRI could be beneficial. The algorithm first considers age and gram stain results as initial screening criteria. Stacchiotti et al. [32] note that the mean age of patients diagnosed with TGCT ranges from 35–50 years on average; subsequently, based on the lower boundary of this range that captured all relevant patients in the present study, patients over 35 years or with a positive gram stain are immediately directed to standard SAH protocol. For patients ≤ 35 years with negative gram stains, the algorithm then considers five additional criteria: female gender, CRP < 125 mg/L, blood WBC < 11,000 cells/μL, afebrile status (temperature < 38 °C), and hip synovial fluid WBC count < 50,000 cells/mL. If three or more of these criteria are met, MRI is recommended.

Fig. 3.

Proposed diagnostic algorithm to identify patients who may benefit from preoperative MRI to differentiate benign intra-articular tumors from septic arthritis of the hip (SAH). SAH: septic arthritis of the hip; CRP: C-reactive protein levels (mg/L); WBC: synovial white blood cell counts (cells/mL); MRI: magnetic resonance imaging. Note: This proposed algorithm is preliminary and requires validation in larger, prospective studies before clinical implementation. Clinical judgment should always be used in conjunction with this tool

The age criterion of < 35 years is particularly effective in distinguishing between the groups, as all tumor cases in our study fell within this range. This aligns with the known epidemiology of TGCT, which typically affects younger individuals compared to SAH [1, 23, 35]. Additionally, gender is included as a criterion due to the strong female predominance (n = 6/7, 85.7%) observed in our tumor group. While the gender difference between the tumor and non-tumor groups only approaches statistical significance in our small sample (p = 0.094), the authors believe that this reflects a trend that warrants consideration and increased clinical suspicion in this patient population. The CRP threshold of 125 mg/L was determined through Receiver Operating Characteristic (ROC) curve analysis, which showed optimal discrimination between tumor and non-tumor cases. The blood WBC count threshold of 11,000 cells/μL, temperature threshold of 38 °C, and synovial fluid WBC count threshold of 50,000 cells/mL were chosen based on established parameters for infection and inflammation.

When applied to our study population, the algorithm demonstrated 100% sensitivity for tumor cases (7/7) and 87.5% specificity for non-tumor cases (14/16 correctly excluded). The algorithm excluded 10 non-tumor cases based on age > 35 years alone, and an additional 2 cases due to positive gram stain. Among the remaining 4 patients ≤ 35 years with negative gram stain, 2 were correctly excluded and 2 were incorrectly included based on meeting at least three of the five additional criteria. It would have recommended MRI for all tumor cases, including the male patient, while excluding most non-tumor cases.

This algorithm could potentially lead to earlier detection of benign intra-articular tumors, reducing misdiagnosis and unnecessary antibiotic treatment. It may also help prioritize cases for urgent MRI in settings with limited immediate access. Future research is needed to validate and refine this algorithm across diverse patient populations as well as to assess its impact on clinical outcomes and resource utilization.

Limitations

This study is not without limitations. Firstly, the retrospective design and limited sample size restrict the generalizability of the findings and highlight the need for larger, prospective studies to validate these results. The small sample size also limits the statistical power of our analyses, potentially masking clinically relevant differences between groups. Secondly, while our study includes a comparison group of confirmed SAH cases, the small sample size in both groups may limit the robustness of our comparative analyses. Future studies with larger cohorts could provide more definitive comparisons of diagnostic markers between benign intra-articular tumors and SAH. Thirdly, the inclusion of only patients who underwent arthroscopic surgery may introduce a selection bias, as patients with milder presentations or those managed non-operatively were not included. This bias may affect the generalizability of the study findings to the broader population of patients with suspected SAH. Additionally, the single-center nature of the study may limit the external validity of the results, as diagnostic and treatment practices may vary across institutions. Finally, while our proposed diagnostic algorithm shows promise, it requires validation in larger, more diverse patient populations before it can be recommended for widespread clinical use. Future prospective studies should focus on refining this algorithm and assessing its impact on clinical outcomes and resource utilization.

Despite these limitations, this study provides valuable insights into the diagnostic challenges of differentiating benign intra-articular tumors from septic arthritis of the hip and offers a foundation for future research in this area.

Conclusions

Differentiating benign intra-articular tumors, particularly TGCT, from SAH poses a diagnostic challenge, especially given histopathologic evidence of tumor pedicle torsion and resultant necrosis that may explain TGCT's acute presentation mimicking SAH. Including benign tumors in the differential diagnosis for suspected SAH is critical, especially among younger and female patients. MRI can be valuable in identifying intra-articular lesions when clinical and laboratory findings are equivocal. The proposed diagnostic algorithm offers a framework for identifying cases where preoperative MRI may be beneficial, though it requires validation in larger studies.

Abbreviations

CRP

C-reactive protein

MRI

Magnetic resonance imaging

SAH

Septic arthritis of the hip

TGCT

Tenosynovial giant cell tumor

WBC

White blood cell

Authors’ contributions

All authors contributed to the formulation of the work. Major contributions are as follows: ND: conception, design, acquisition, analysis, interpretation; ER: conception, design, interpretations, substantial revisions; IA: acquisition, analysis, interpretation, substantial revisions; MD: acquisition, analysis, interpretation, substantial revisions; OS: acquisition, analysis, interpretation, substantial revisions; AS: analysis, interpretation, substantial revisions, corresponding author responsibilities; NP: interpretation, substantial revision; EA: conception, design, interpretations, substantial revisions. All authors have approved the submitted version of the manuscript and are accountable to the accuracy and integrity of the study.

Funding

Not applicable.

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files.

Declarations

Ethics approval and consent to participate

Institutional review board approval was sought and obtained prior to study initiation at the participating institution’s IRB board (Israel Ministry of Health, Tel Aviv Sourasky Medical Center, Affiliated to the Tel Aviv University Sackler Faculty of Medicine Municipality of Tel Aviv-Yaffo) with IRB number TLV-0402–13 and the study was conducted in accordance with the Declaration of Helsinki and signed by the director of the medical institution and chairman of the Helsinki Committee of the Medical Institution. All patients consented to study participation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.