Xanthogranulomatous cholecystitis mimicking gallbladder carcinoma

Abstract

Xanthogranulomatous cholecystitis (XGC) is a rare benign inflammatory disorder that can closely resemble gallbladder carcinoma (GBC) clinically, radiologically, intraoperatively, and even on gross pathological examination. This overlap may lead to both unnecessary radical surgery and undertreatment of unsuspected malignancy. This review summarizes the contemporary literature together with institutional clinical and radiological experience, focusing on diagnostic pitfalls and practical surgical decision-making. Two illustrative cases are also presented. Reliable preoperative differentiation between XGC and GBC remains difficult because symptoms, laboratory findings, tumor markers, and individual imaging features overlap considerably. A pattern-based interpretation of computed tomography and magnetic resonance imaging findings may improve diagnostic confidence. Diffuse wall thickening, continuous mucosal enhancement, intramural hypoattenuating or T2-hyperintense nodules, gallstones, and preservation of the mural architecture tend to favor XGC, whereas focal mass formation, mucosal disruption, biliary obstruction, lymphadenopathy, and convincing hepatic invasion raise greater concern for GBC. Machine-learning and deep-learning models have shown promising diagnostic performance, although external validation and improved interpretability are still required before they can be incorporated into routine clinical practice. Frozen-section analysis, careful nodal assessment, and delayed oncologic resection after definitive histological assessment when uncertainty persists may provide a reasonable balance between avoiding unnecessary radical procedures and preventing undertreatment of true malignancy.

Key Words: Xanthogranulomatous cholecystitis; Gallbladder carcinoma; Cholecystectomy; Surgical decision-making; Surgery

Core Tip: Xanthogranulomatous cholecystitis is a benign inflammatory disease that may closely mimic gallbladder carcinoma, creating a risk of either unnecessary radical surgery or delayed cancer treatment. No single clinical, laboratory, or imaging feature reliably separates the two conditions. Diagnostic confidence improves when clinical findings and multimodal imaging features are interpreted. Frozen-section analysis, nodal assessment, safe cholecystectomy with early conversion or bail-out procedures when necessary and delayed oncologic resection after definitive histological assessment in unresolved cases may offer the best compromise between oncologic adequacy and avoidance of overtreatment.

INTRODUCTION

Gallbladder carcinoma (GBC) is an uncommon but highly aggressive biliary tract malignancy, with substantial geographic, ethnic, and sex-related variation[1,2]. The highest incidence rates have been reported in South America, northern India, Eastern Europe, and selected East Asian populations, whereas the disease is less common in most Western countries. GBC occurs more frequently in women and in older individuals, a distribution that partly reflects the prevalence of major risk factors such as gallstone disease, chronic gallbladder inflammation, obesity, hormonal influences, and genetic or environmental susceptibility[1-4].

Xanthogranulomatous cholecystitis (XGC) is a rare inflammatory variant of chronic cholecystitis, identified in approximately 1%-9% of cholecystectomy specimens[5-7]. Although benign, it has considerable clinical relevance because its radiological and intraoperative appearance may be virtually indistinguishable from GBC. In particular, the inflammatory process can extend into adjacent tissues and create the appearance of locally invasive malignancy.

This diagnostic overlap poses an important therapeutic dilemma. The challenge is not simply to distinguish XGC from GBC, but to translate diagnostic uncertainty into a safe and oncologically appropriate surgical strategy. This minireview therefore focuses on contemporary pattern-based imaging assessment, emerging predictive models, and intraoperative approaches designed to avoid both unnecessary radical surgery for benign disease and inadequate treatment of unsuspected carcinoma. Two illustrative cases are included to show how these challenges arise in clinical practice.

PATHOGENESIS AND RISK FACTORS

The development of GBC is closely linked to chronic epithelial injury and persistent inflammation of the gallbladder mucosa. Cholelithiasis is the most consistently reported risk factor, and gallstones are present in most patients with GBC[8,9]. Nevertheless, only a small proportion of individuals with gallstones develop carcinoma, indicating that malignant transformation depends on additional biological, environmental, and host-related factors. Large gallstones, longstanding inflammation, porcelain gallbladder, gallbladder polyps with high-risk features, anomalous pancreatobiliary duct junction, chronic infections such as Salmonella typhi carriage, obesity, and primary sclerosing cholangitis have all been associated with increased risk[1,4,10,11].

The proposed carcinogenic sequence involves chronic irritation and mucosal injury, followed by dysplasia and eventual progression to invasive adenocarcinoma. Reproductive and hormonal factors may contribute to the female predominance of the disease, while ethnic clustering and familial aggregation suggest a role for genetic susceptibility in selected populations[10,12-14].

The pathogenesis of XGC is different. The disease is thought to result from extravasation of bile into the gallbladder wall, usually in the setting of gallstones, cystic duct obstruction, or acute or chronic inflammation. Rupture of Rokitansky-Aschoff sinuses or mucosal ulceration allows bile and cholesterol pigments to enter the gallbladder wall, where they trigger an intense histiocytic inflammatory response. Foamy macrophages, multinucleated giant cells, cholesterol clefts, lymphocytes, plasma cells, and fibrosis accumulate within the wall, ultimately producing a mass-like inflammatory lesion[15,16].

This destructive inflammatory process also explains the tendency of XGC to extend into surrounding tissues, form dense adhesions, and mimic locally advanced malignancy (Figure 1). Although XGC itself is not regarded as a premalignant condition, coexistence with GBC has been described in a minority of patients. A diagnosis of XGC should therefore not automatically be considered sufficient to exclude malignancy, particularly when imaging or operative findings remain suspicious.

Figure 1 Histologic features of xanthogranulomatous cholecystitis. A: Low-power view showing an irregularly and unevenly thickened gallbladder wall in a case of xanthogranulomatous cholecystitis; B: The perimuscular soft tissues show a dense inflammatory infiltrate predominantly composed of numerous foamy histiocytes and lymphocytes. No Michaelis-Gutmann bodies, which are characteristic of malakoplakia, are identified; C: In addition to the inflammatory infiltrate, fibroblasts and fibrotic tissue displaying a vaguely storiform pattern are present (right side), potentially posing diagnostic challenges with neoplastic diseases; D: CD68 (PGM1) immunohistochemical staining highlights the predominant histiocytic component.

CLINICAL PRESENTATION AND LABORATORY FINDINGS

The clinical presentation of XGC and GBC is often nonspecific, and there is considerable overlap between the two conditions. Both may present with right upper quadrant pain, nausea, vomiting, fever, anorexia, weight loss, jaundice, or symptoms of acute or chronic cholecystitis[4,11,17].

XGC more commonly presents as inflammatory gallbladder disease, with abdominal pain, fever, leukocytosis, elevated C-reactive protein, and recurrent episodes of cholecystitis. Murphy’s sign may be positive, and complications such as empyema, Mirizzi syndrome, biliary fistula, choledocholithiasis, or pancreatitis can occur. Weight loss is also possible and may further complicate the differential diagnosis by raising suspicion of malignancy[18-20].

Early-stage GBC is frequently asymptomatic or presents with symptoms indistinguishable from those of benign biliary disease. As a result, many tumors are diagnosed incidentally after cholecystectomy performed for presumed gallstones or cholecystitis. Jaundice, anorexia, persistent pain, a palpable mass, or substantial weight loss are more often associated with advanced disease and should prompt careful staging.

Laboratory investigations may support clinical assessment but are rarely decisive. XGC is often accompanied by elevated inflammatory markers, including leukocytosis and increased C-reactive protein levels. Liver function tests may be abnormal in both diseases, particularly when biliary obstruction is present. Tumor markers such as carbohydrate antigen 19-9 (CA 19-9) and carcinoembryonic antigen (CEA) also have limited specificity. Although CA 19-9 may be elevated in GBC, marked increases can occur in benign inflammatory or obstructive biliary conditions, including XGC[19,21-23]. Tumor marker results must therefore be interpreted within the wider clinical, radiological, and, when available, pathological context.

PREOPERATIVE DIAGNOSTIC WORK-UP

The main aim of the preoperative work-up is to estimate the likelihood of malignancy and determine the most appropriate surgical strategy. In practice, however, reliably distinguishing XGC from GBC before surgery remains difficult because no individual clinical, laboratory, or imaging finding provides sufficient discriminatory accuracy. Assessment is therefore best based on a combination of clinical evaluation, laboratory investigations, and cross-sectional imaging within a multimodal framework[24,25].

Tumor markers

Routine laboratory investigations provide limited discriminatory information. Patients with XGC may show a more pronounced systemic inflammatory response than those with GBC, but leukocyte count, liver function tests, and bilirubin levels overlap considerably between the two groups[26]. Serum tumor markers present a similar problem. CA 19-9, while commonly elevated in GBC, can also reach high levels in XGC, acute cholecystitis, cholangitis, and obstructive jaundice, substantially limiting its specificity for malignancy[19,24,27]. Elevated CEA may increase suspicion for GBC, but normal values do not reliably exclude carcinoma[19,26].

The largest comparative cohort currently available, which included 240 patients, 90 with XGC and 150 with GBC, identified CEA as the only serum biomarker significantly associated with malignancy. Alpha-fetoprotein, cancer antigen 125, and carbohydrate antigen 242 had limited discriminatory value[28]. Subsequent studies have likewise shown considerable overlap in conventional tumor marker levels between the two diseases. Even the most informative markers therefore lack the accuracy required for independent clinical decision-making[26,28].

More recently, systemic inflammatory biomarkers have been explored as potential adjuncts to preoperative risk stratification. One example is the systemic immune-inflammation index (SII), calculated as platelet count × neutrophil count/Lymphocyte count, which may reflect the interaction between systemic inflammation and tumor-associated immune responses. In a retrospective study of 33 patients with XGC and 22 with GBC, an SII cutoff of 640 provided a sensitivity of 77.3%, specificity of 66.7%, negative predictive value of 81.5%, and overall diagnostic accuracy of 79.9%[29]. These findings suggest that inflammatory indices may provide information beyond conventional tumor markers, although the available evidence remains preliminary and requires prospective validation.

Overall, laboratory biomarkers play a complementary rather than decisive role in differentiating XGC from GBC. CEA and inflammatory indices such as SII may contribute to preoperative risk stratification, but neither has sufficient accuracy to be used as a standalone diagnostic test. Their value lies primarily in combination with clinical and imaging findings[26,28,29].

Imaging features and emerging diagnostic models

The main difficulty in imaging is not the absence of potentially useful radiological signs, but rather their overlap between inflammatory and malignant disease. Several features have been associated with XGC or GBC, yet few are sufficiently specific when considered in isolation. Diagnostic interpretation is therefore more reliable when multiple findings are considered together. This pattern-based approach has increasingly replaced reliance on single radiological signs and has provided the conceptual basis for more structured multiparametric models[24,30].

Conventional imaging and sources of diagnostic overlap

Ultrasonography is generally the first imaging examination performed in patients with gallbladder disease. XGC typically manifests as diffuse wall thickening, intramural hypoechoic nodules, and cholelithiasis. In contrast, GBC more commonly presents as a focal intraluminal mass, asymmetric wall thickening, a sessile polypoid lesion, or replacement of the gallbladder lumen by heterogeneous soft tissue[24,31]. However, ultrasonography alone rarely provides sufficient information to establish the diagnosis, particularly because assessment of local invasion, biliary involvement, and regional lymphadenopathy is limited.

Contrast-enhanced computed tomography (CT) is the most extensively studied imaging modality for differentiating XGC from GBC. Features most consistently associated with XGC include diffuse wall thickening, continuous mucosal enhancement, intramural hypoattenuating nodules, and the absence of convincing hepatic invasion or intrahepatic biliary dilatation. Their diagnostic value is greatest when several findings are present simultaneously. Goshima et al[30] reported that the presence of at least three of five characteristic CT findings provided a sensitivity of 83%, specificity of 100%, and overall diagnostic accuracy of 91%. Subsequent comparative studies have broadly supported these observations, linking preserved mural architecture and intramural nodules with XGC, while focal mass formation, mucosal disruption, biliary obstruction, regional lymphadenopathy, and convincing hepatic invasion are more suggestive of GBC[26,28,32,33].

Even so, overlap remains substantial. Severe xanthogranulomatous inflammation can extend beyond the gallbladder wall and produce apparent hepatic or hilar infiltration, vascular involvement, biliary obstruction, and reactive lymphadenopathy, closely resembling locally advanced carcinoma. Conversely, some malignant lesions may retain features more commonly associated with inflammatory disease. CT findings are therefore best interpreted as a constellation rather than as isolated diagnostic criteria. The most relevant features and their major areas of overlap are summarized in Table 1.

Table 1 Key imaging findings, diagnostic interpretation, and sources of overlap between xanthogranulomatous cholecystitis and gallbladder carcinoma.

Imaging finding	Typical interpretation	Relative diagnostic value	Why it favors this diagnosis	Source of diagnostic overlap
Diffuse wall thickening	XGC	Moderate	Chronic transmural inflammation	May occur in infiltrative GBC
Continuous mucosal line	XGC	Higher	Relative preservation of mucosal integrity	Can be partially preserved in early GBC
Intramural hypoattenuating nodules	XGC	Higher	Xanthogranulomatous inflammation	Not universally present
Gallstones	XGC	Limited	Chronic obstruction and inflammation	Common in both diseases
Intraluminal mass	GBC	Moderate	Neoplastic tissue proliferation	Inflammatory pseudotumoral lesions may mimic mass formation
Lymphadenopathy	GBC	Moderate	Regional tumor spread	Reactive inflammatory nodes may occur in XGC
Biliary obstruction	GBC	Moderate	Ductal invasion or compression	Severe inflammatory disease may produce similar findings
Hepatic invasion	GBC	Higher	Direct local extension	Inflammatory infiltration may mimic true invasion
Vascular encasement	GBC	Higher	Advanced locoregional disease	Rare inflammatory pseudoinvasion

Relative diagnostic value is intended as a qualitative assessment based on the consistency and discriminatory relevance of each finding across the studies discussed in this review; it does not represent a formal grading of the level of evidence. XGC: Xanthogranulomatous cholecystitis; GBC: Gallbladder carcinoma.

Magnetic resonance imaging (MRI) can provide further information when CT findings are equivocal, particularly through improved characterization of mural architecture, tissue composition, enhancement patterns, and diffusion characteristics. Diffuse wall thickening, preservation of mural architecture, intramural T2-hyperintense nodules, gallstones, and delayed fibrotic enhancement tend to support XGC, whereas focal irregular thickening, infiltrative growth, and marked diffusion restriction increase suspicion for GBC[34]. These distinctions are not absolute: Dense inflammation and fibrosis in XGC may also cause diffusion restriction and irregular enhancement. As with CT, MRI should therefore be interpreted using a combination of findings rather than any single feature (Table 1).

Structured multiparametric assessment has been explored as one way of improving diagnostic accuracy. In a retrospective cohort of 23 patients with XGC and 35 with wall-thickening-type GBC, Han et al[34] developed an MRI scoring system incorporating nine imaging variables, including diffuse wall thickening, mucosal uniformity, intramural T2-high signal intensity, mucosal retraction, gallstones, diffusion restriction, enhancement pattern, and phase of peak enhancement. The resulting model achieved an area under the curve (AUC) of 0.972 and outperformed conventional visual assessment by radiologists. These results are encouraging, although the small retrospective derivation cohort limits their generalizability and prospective external validation remains necessary.

Complementary diagnostic modalities

Endoscopic ultrasound may provide useful information on mural stratification, local invasion, and regional lymph nodes and can allow tissue sampling in selected cases. Its reliability is nevertheless limited by sampling error and by the possibility that XGC and GBC may coexist within the same gallbladder[24,27]. Positron emission tomography is similarly of limited value for differential diagnosis because inflammatory lesions may show substantial metabolic uptake. Its main role is therefore as an adjunct to staging rather than as a tool for distinguishing inflammation from malignancy[24,31].

Emerging predictive models

The persistent diagnostic overlap between XGC and GBC has prompted increasing interest in computational approaches capable of integrating clinical, laboratory, and imaging information within a single predictive model[28]. Fu et al[35] developed a multicenter machine-learning nomogram based on sex, Murphy’s sign, absolute neutrophil count, gamma-glutamyl transferase, CEA, and comprehensive imaging diagnosis. The model showed high discriminatory performance in both the development and external validation cohorts.

More recently, a large multicenter retrospective study involving 1246 patients developed an integrated machine-learning model combining clinical, laboratory, and imaging variables. The model achieved an AUC of 0.94 during internal validation and 0.88 in external testing and also performed favorably compared with radiologist assessment[36]. These findings reinforce the potential value of multimodal computational approaches in the preoperative distinction between XGC and GBC.

Their clinical translation, however, remains uncertain. Most models have been developed retrospectively and are therefore vulnerable to selection bias, differences in case mix, and optimistic estimates of diagnostic performance. External validation has been reported for some models, but validation cohorts are often drawn from geographically related institutions and may share similar referral patterns, imaging protocols, and population characteristics. Performance may consequently decline when these models are applied in centers with different disease prevalence, scanner platforms, acquisition protocols, or radiological expertise (Table 2).

Table 2 Representative diagnostic and predictive models developed for the differentiation of xanthogranulomatous cholecystitis and gallbladder carcinoma.

Ref.	Study design/country	Diagnostic framework	Population	External validation	Performance	Principal limitations
Goshima et al[30], 2010	Retrospective, single-center/Japan	CT-based pattern assessment using five imaging findings	18 XGC/17 GBC	No	Accuracy 91% when ≥ 3 characteristic CT findings were present	Very small retrospective cohort; CT-based assessment only; no external validation
Xiao et al[28], 2022	Retrospective, single-center/China	Integrated clinical, laboratory, and radiological assessment	90 XGC/150 GBC	No	Mucosal line continuity and low-density border showed the highest discriminatory value	Single-center retrospective design; no independent validation; limited performance of individual features
Han et al[34], 2024	Retrospective, single-center/South Korea	Multiparametric magnetic resonance imaging scoring system	23 XGC/35 GBC	No	AUC 0.972	Small selected cohort restricted to wall-thickening-type GBC; no external validation
Fu et al[35], 2024	Retrospective, two-center/China	Machine-learning-assisted clinical, laboratory, and imaging nomogram	290 development/146 external validation	Yes, independent second center	Development AUC 0.936; consistent performance in external validation	Retrospective design; geographically restricted population; limited prospective clinical evaluation
Zhang et al[36], 2025	Retrospective, four-center/China	LIDGAX multimodal machine-learning model integrating clinical, laboratory, and imaging variables	1246 total; 674 training, 169 internal validation, 279 external testing, and 124 real-world evaluation	Yes, independent multicenter external testing	AUC 0.94 internal validation; 0.88 external testing; 0.95 real-world evaluation	Retrospective design; all centers from a single country/region; dependence on multimodal imaging availability; prospective impact on surgical outcomes not yet established

CT: Computed tomography; XGC: Xanthogranulomatous cholecystitis; GBC: Gallbladder carcinoma; AUC: Area under the curve.

Moreover, a high AUC does not necessarily imply that a model improves clinical care. There is currently little evidence that using these tools reduces unnecessary radical resections, prevents undertreatment of unsuspected GBC, or improves patient outcomes compared with expert multidisciplinary assessment. Limited interpretability, particularly in deep-learning systems, may also restrict clinician confidence and make individual predictions difficult to explain.

Future studies should therefore focus not only on discrimination but also on prospective multicenter validation, standardized imaging acquisition, calibration, transparency, and explainability. Most importantly, predictive models should be tested for their incremental value in real-world surgical decision-making.

SURGICAL MANAGEMENT AND OUTCOMES

The principles of surgical treatment for XGC and GBC are fundamentally different. In practice, however, management can become difficult when preoperative imaging and operative findings do not allow the two conditions to be reliably distinguished. For suspected or confirmed GBC, the goal is oncologic radicality, with achievement of an R0 resection remaining one of the strongest determinants of long-term survival. Simple cholecystectomy is generally considered adequate for T1a disease, whereas more advanced tumors usually require radical cholecystectomy with hepatic resection and regional lymphadenectomy. Liver resection may consist of a wedge excision of the gallbladder bed or anatomical resection of segments S4b and S5, with the extent tailored to tumor stage, depth of hepatic invasion, anatomical involvement, and the likelihood of achieving negative margins while preserving adequate liver function[37].

Major hepatectomy should be reserved for selected patients in whom it is necessary to obtain an R0 resection, given the higher morbidity associated with the procedure and the absence of a consistent survival advantage when it is performed unnecessarily[38]. Regional lymphadenectomy remains essential for staging and prognostic assessment, and retrieval of at least six lymph nodes is generally recommended[37,39]. Routine bile duct resection is not indicated and should usually be limited to cases of direct biliary involvement or a positive cystic duct margin[40-42].

The role of minimally invasive surgery in GBC remains selective. Laparoscopic or robotic radical cholecystectomy may offer perioperative advantages in carefully selected patients with early-stage tumors when performed in experienced hepatopancreatobiliary centers. Open surgery remains preferable when locally advanced disease, extensive hepatic invasion, vascular involvement, or multivisceral resection is anticipated. Whatever the approach, careful manipulation of the gallbladder is essential to minimize the risk of perforation, bile spillage, and potential tumor dissemination[42].

In presumed XGC, surgery has both therapeutic and diagnostic purposes. Cholecystectomy treats the underlying inflammatory disease and provides definitive histopathological confirmation, which remains important because XGC may closely resemble GBC before and during surgery. Although a laparoscopic approach can be attempted, the operation is often technically demanding because of dense adhesions, severe fibrosis, marked wall thickening, and distortion or obliteration of Calot’s triangle.

Conversion to open surgery should not be viewed as a procedural failure in this setting. Rather, it represents an appropriate safety measure when the critical view of safety cannot be obtained, anatomy remains unclear, significant bleeding occurs, or concern for malignancy persists. In particularly difficult cases, subtotal cholecystectomy is an important bail-out option that may avoid hazardous dissection in Calot’s triangle. Depending on the technique used, however, it carries a risk of bile leakage, retained stones, and recurrent biliary symptoms[20,24,43-45].

The greatest surgical challenge arises when XGC and GBC remain indistinguishable intraoperatively. Both can present with pronounced wall thickening, hard mass-like lesions, dense adhesions to adjacent structures, apparent liver infiltration, and enlarged regional lymph nodes. Macroscopic appearance alone is therefore often insufficient to distinguish advanced inflammatory disease from neoplastic invasion.

Frozen-section analysis can be particularly useful in this setting and may prevent unnecessary radical surgery when adequate sampling demonstrates benign inflammation. Its limitations must nevertheless be recognized. Extensive fibrosis may complicate pathological interpretation, sampling error is possible, and small foci of carcinoma may coexist with xanthogranulomatous inflammation. A negative frozen section should therefore be interpreted in the context of sampling adequacy and the overall level of macroscopic suspicion[21,26,46].

The intraoperative strategy should balance oncologic adequacy against the morbidity of unnecessary extended resection. When malignancy is confirmed and an R0 resection appears achievable, radical cholecystectomy with the appropriate hepatic resection and lymphadenectomy should be pursued. In contrast, when frozen section supports XGC and there is no convincing residual evidence of malignancy, unnecessary oncologic extension should be avoided.

If uncertainty remains and an extended procedure would expose the patient to substantial additional morbidity, completing a safe cholecystectomy and considering delayed oncologic resection after definitive histopathological assessment may be preferable. The key principle is that diagnostic uncertainty should lead to oncologic preparedness, not automatic radicalization. Frozen-section analysis, careful assessment of regional nodes, avoidance of gallbladder perforation, and delayed oncologic resection when appropriate can help reduce both overtreatment of benign XGC and undertreatment of true carcinoma.

A practical approach can therefore be summarized according to the clinical scenario. Suspected early-stage GBC should be treated with oncologic resection according to stage. Suspected locally advanced GBC requires careful staging and avoidance of futile R1 or R2 surgery. Presumed XGC should be managed with safe cholecystectomy, with early conversion or subtotal cholecystectomy when anatomy is unsafe. When XGC mimics GBC and the diagnosis remains uncertain, frozen section, nodal assessment, and, where appropriate, delayed oncologic resection after definitive histological assessment provide a reasonable strategy[38,41].

Outcomes

Surgical outcomes in XGC are generally favorable, although perioperative morbidity is not negligible because the operation itself can be technically demanding. Operative time, blood loss, conversion rates, postoperative morbidity, and length of stay vary substantially among published series. This variability reflects differences in disease severity, timing of surgery, surgical approach, surgeon experience, and the need for additional procedures such as ERCP or percutaneous drainage.

Dense fibrosis, inflammatory adhesions, distortion of Calot’s triangle, and extension into the liver bed are the main factors contributing to operative difficulty. Reported postoperative complications include surgical-site infection, bile leakage, pneumonia, urinary tract infection, and wound complications, particularly among patients requiring open conversion or additional procedures. Despite this technical complexity, postoperative mortality remains very low when sound surgical judgment is applied[7,20,24,43,44,47].

In XGC, adverse perioperative outcomes are driven mainly by operative complexity and diagnostic confusion rather than by malignant biology. This creates two opposing risks. Benign XGC may be mistaken for locally advanced GBC, resulting in unnecessary hepatic, biliary, or multivisceral resection. Conversely, carcinoma may be interpreted as inflammatory disease, leading to inadequate initial treatment or delayed oncologic management. Misdiagnosis therefore has direct implications not only for diagnostic accuracy but also for the extent of surgery, perioperative morbidity, and the likelihood of requiring reoperation.

Long-term outcomes differ substantially between XGC and GBC. Complete cholecystectomy is generally curative for XGC, and recurrence after complete excision is rarely reported. After subtotal cholecystectomy, follow-up may be appropriate because of the risk of recurrent symptoms arising from the gallbladder remnant. In GBC, by contrast, prognosis is largely determined by tumor stage, nodal status, vascular or distant spread, and completeness of resection. R0 resection is associated with improved survival, whereas R1 and R2 resections are associated with high recurrence rates. Thus, outcomes in XGC are primarily related to inflammation-driven technical difficulty, while those in GBC are governed by tumor biology and oncologic clearance[48].

Overall, the management of XGC mimicking GBC requires a careful compromise between surgical safety and oncologic adequacy. Surgeons should be prepared to follow oncologic principles whenever malignancy is suspected, while avoiding automatic extension of the procedure in the absence of adequate pathological or anatomical evidence. Meticulous dissection, timely conversion, bail-out procedures for unsafe anatomy, frozen-section analysis, nodal assessment, and delayed oncologic resection in selected patients provide a pragmatic framework for minimizing both overtreatment and undertreatment[24,38].

CASE SERIES

The following two cases are not intended to represent a formal clinical series. Rather, they illustrate how the diagnostic uncertainty described above can directly influence operative decision-making. Both patients had radiological and intraoperative findings that raised substantial concern for malignancy, yet careful reassessment, selective use of frozen section, and avoidance of automatic radical resection ultimately prevented unnecessary extended surgery.

Case 1

A 52-year-old man with cholelithiasis was referred for suspected GBC after an episode of acute abdominal pain. CT showed marked irregular thickening of the gallbladder wall, reaching 18-19 mm, together with pericholecystic fluid, loss of the cleavage plane with segment S5, and a hilar lymph node considered reactive. CA 19-9 was elevated at 208 U/mL. MRI confirmed marked wall thickening with diffusion restriction and poor demarcation from the adjacent liver, further increasing suspicion of malignancy (Figure 2).

Figure 2 Multimodal preoperative imaging of xanthogranulomatous cholecystitis mimicking gallbladder carcinoma. A and B: Contrast-enhanced computed tomography in the portal venous and arterial phases showing marked irregular thickening of the gallbladder wall and loss of the cleavage plane with the adjacent liver parenchyma; C: Diffusion-weighted magnetic resonance imaging demonstrating marked diffusion restriction in the gallbladder body-fundus region, further raising suspicion for a neoplastic lesion; D: T2-weighted magnetic resonance imaging showing abnormal gallbladder wall thickening and pericholecystic inflammatory changes, with poor demarcation from the adjacent hepatic parenchyma.

Robotic exploration revealed dense omento-hepatic adhesions, an inflamed gallbladder embedded in the liver bed, and several small whitish peritoneal nodules. Because disseminated malignancy could not be excluded, peritoneal fluid, falciform nodules, and a Glissonian nodule were submitted for cytology and frozen-section analysis. All were negative for malignancy. Frozen section of the gallbladder fundus was likewise negative. An anterograde cholecystectomy was therefore completed without hepatic radicalization. The postoperative course was uneventful, and final pathology confirmed XGC without neoplastic cells.

Case 2

A 73-year-old man was evaluated for asthenia and malaise. Ultrasound demonstrated biliary sludge and focal vascularized thickening of the gallbladder fundus. CT confirmed irregular enhancing mural thickening associated with a solid exophytic lesion measuring approximately 19 mm × 11 mm and loss of the cleavage plane with segments S4b and S5, raising concern for hepatic invasion. Contrast-enhanced ultrasound also demonstrated a vascularized vegetating lesion, and radical cholecystectomy with planned S4b-S5 resection was scheduled.

At robotic exploration, there was no evidence of peritoneal disease or distant spread. The gallbladder was firm, hypervascular, markedly inflamed, and densely adherent to the omentum and abdominal wall. Calot’s triangle was severely distorted and the critical view of safety could not be obtained, prompting conversion to open surgery. Intraoperative reassessment did not demonstrate convincing hepatic invasion or suspicious lymphadenopathy. Liver resection was therefore omitted, and an anterograde cholecystectomy was performed. Final histopathology confirmed XGC without malignancy.

Together, these cases illustrate the main practical message of this review. In both patients, preoperative imaging strongly suggested locally invasive GBC, yet the operative findings remained compatible with severe inflammatory disease. The first case highlights the usefulness of targeted frozen-section analysis when disseminated or locally advanced malignancy is suspected. The second emphasizes the importance of reassessing the need for extended resection intraoperatively when direct hepatic invasion is not convincingly demonstrated.

STRENGTHS AND LIMITATIONS OF THE CURRENT EVIDENCE

Despite the limitations of the available literature, several clinically useful observations emerge consistently across independent studies. In particular, combinations of imaging features appear more informative than individual signs. Preservation of the mucosal line, diffuse wall thickening, intramural nodules, and gallstones have repeatedly been associated with XGC, supporting a pattern-based approach to diagnosis. The increasing integration of clinical, laboratory, and imaging variables into structured multiparametric and computational models is another important development, moving the field beyond conventional qualitative image interpretation.

At the same time, the evidence base remains relatively weak. Most published studies are retrospective, include small or highly selected cohorts, and originate from tertiary referral centers. Differences in patient selection, imaging protocols, radiological definitions, and reference standards make direct comparisons difficult and may limit generalizability to routine practice. In addition, studies designed specifically to distinguish XGC from GBC are necessarily enriched for diagnostically difficult cases and may not reflect the broader spectrum of gallbladder abnormalities encountered in unselected clinical populations.

Machine-learning and deep-learning models provide encouraging evidence that multimodal integration may improve diagnostic discrimination. However, as discussed above, strong diagnostic performance does not necessarily translate into clinical benefit. Prospective evidence showing that these models improve surgical decision-making or patient outcomes is still lacking. Future studies should therefore assess whether computational tools add meaningful value to expert multidisciplinary evaluation, particularly by reducing unnecessary radical surgery without increasing the risk of undertreating occult malignancy.

A further limitation is inherent to the biology of the two diseases rather than to the diagnostic techniques themselves. Severe xanthogranulomatous inflammation can reproduce many of the morphological and infiltrative features of carcinoma, and XGC and GBC may occasionally coexist in the same gallbladder. Complete diagnostic certainty may therefore remain difficult to achieve even with increasingly sophisticated imaging and computational tools.

Future studies should prioritize prospective multicenter validation, standardized imaging definitions and acquisition protocols, and clinically interpretable multimodal diagnostic strategies. Until stronger evidence is available, the most reliable approach remains the integration of preoperative imaging, multidisciplinary assessment, and careful intraoperative judgment.

FUTURE CLINICAL PERSPECTIVES

Future research should move beyond conventional morphological assessment and explore integrated, non-invasive strategies capable of improving the preoperative distinction between XGC and GBC. One potentially important direction is the identification of circulating molecular biomarkers that reflect inflammatory, immune, or tissue-remodeling pathways. Circulating proteins, cytokine profiles, extracellular vesicles, circulating nucleic acids, and other molecular signatures could complement conventional tumor markers and imaging findings. Their usefulness will need to be established prospectively in well-characterized cohorts that include both inflammatory gallbladder disease and histologically confirmed carcinoma.

Another priority is the development of explainable diagnostic models that combine clinical characteristics, laboratory biomarkers, and multimodal imaging. Future machine-learning and deep-learning models should be judged on more than discriminatory performance alone. Calibration, external generalizability, interpretability, and incremental benefit over expert multidisciplinary assessment will be equally important. Models that can indicate which features contributed most strongly to an individual prediction may be easier for clinicians to interpret and ultimately more suitable for use in real-world surgical decision-making.

A further area of interest is the inflammatory biology of XGC itself. Better characterization of the pathways driving the xanthogranulomatous response could eventually provide a rationale for targeted anti-inflammatory interventions in selected patients. In principle, modulation of inflammation might reduce inflammatory activity, allow interval radiological reassessment, or facilitate subsequent surgery in patients in whom malignancy is considered unlikely.

At present, however, this remains an exploratory concept. Anti-inflammatory treatment should not postpone definitive diagnosis or oncological management when GBC cannot be confidently excluded. Translational research and prospective clinical studies will be required to identify relevant molecular targets, appropriate patient-selection criteria, optimal treatment timing, and clinically meaningful endpoints.

Ultimately, future diagnostic strategies should be assessed not simply according to how accurately they distinguish XGC from GBC, but according to whether they change management in a clinically useful way, most importantly, whether they reduce unnecessary radical resections without increasing the risk of undertreating occult malignancy.

CONCLUSION

XGC remains one of the most challenging benign mimics of GBC. Despite improvements in cross-sectional imaging and the emergence of predictive models, reliable preoperative differentiation remains difficult because inflammatory and malignant findings frequently overlap. Management should therefore combine multimodal assessment with careful intraoperative judgment and oncologic preparedness, while avoiding automatic radicalization. When malignancy is confirmed and an R0 resection is feasible, radical surgery is appropriate. When pathological assessment supports benign inflammation, unnecessary extended resection should be avoided. In cases in which uncertainty persists, delayed oncologic resection after definitive histological confirmation may provide the safest balance between oncologic adequacy and surgical prudence.