Multimodal Diagnosis and Management of Gastric Tuberculosis: A Case Report

Summary

This case report presents a stepwise multimodal workflow for diagnosing gastric tuberculosis, integrating gastroscopy, cross-sectional imaging, endoscopic ultrasound, histopathology, and molecular testing to confirm infection when superficial biopsies are non-diagnostic.

Abstract

Gastric tuberculosis is an exceptionally rare form of extrapulmonary tuberculosis and often presents with non-specific symptoms and endoscopic appearances, frequently mimicking submucosal tumors, peptic ulcer disease, or malignancy. These overlaps, together with the limited yield of superficial biopsies, can delay definitive diagnosis. Here, researchers report a case of gastric tuberculosis and emphasize the methodological advantage of a multimodal diagnostic strategy that integrates endoscopy, cross-sectional imaging, histopathology, and molecular testing to progressively narrow diagnostic uncertainty. Gastroscopy revealed a localized gastric lesion, while imaging provided complementary lesion characterization and suggested possible extra-gastric involvement, prompting further evaluation for granulomatous infection. Repeated endoscopic sampling did not establish a diagnosis, and laparoscopic partial gastrectomy was therefore performed to obtain adequate deep tissue for comprehensive assessment. Histopathological examination demonstrated granulomatous inflammation with caseous necrosis, and molecular testing supported infection with Mycobacterium tuberculosis complex, confirming gastric tuberculosis. The postoperative course was uneventful, and the patient subsequently received standard antituberculous therapy. Symptoms resolved, and no recurrence was observed during long-term follow-up. This case highlights practical diagnostic challenges and provides a reproducible, stepwise approach that may help clinicians suspect, differentiate, and confirm gastric tuberculosis earlier in similar presentations.

Introduction

Gastric tuberculosis is an exceptionally rare manifestation of Mycobacterium tuberculosis infection in the human body¹. Its incidence is extremely low, and even in regions with a high tuberculosis burden, gastric involvement is typically reported as accounting for less than 1% of gastrointestinal tuberculosis cases². This low incidence has been attributed to several local protective factors, including the bactericidal effect of gastric acid, the integrity of the gastric mucosal barrier, and rapid gastric emptying, which collectively reduce sustained bacillary contact with the gastric wall³.

The clinical manifestations of gastric tuberculosis are heterogeneous and lack pathognomonic features, with symptoms and signs varying by lesion location, depth of involvement, and disease progression⁴. Patients may present with upper abdominal pain or discomfort, nausea, vomiting, anorexia, and weight loss⁵. When gastric involvement coexists with extra-gastric tuberculosis, constitutional symptoms such as low-grade fever, night sweats, and fatigue may also occur⁶. However, these manifestations are non-specific, and clinical suspicion is often delayed because the disease can resemble common gastric conditions, including peptic ulcer disease, gastric carcinoma, lymphoma, and submucosal tumors⁷.

A major diagnostic barrier described in the wider body of literature is that routine endoscopic forceps biopsy is frequently non-diagnostic, particularly when lesions are predominantly submucosal or when granulomatous changes are patchy, leading to sampling error⁸. To overcome this, the multimodal diagnostic strategy presented here provides significant advantages over conventional single-modality endoscopic assessments. By integrating cross-sectional imaging, deep tissue acquisition, and molecular testing (PCR), this comprehensive approach significantly increases diagnostic confidence and minimizes delays when conventional acid-fast stains are negative⁹.

The overall goal of this protocol is to formalize a reproducible, stepwise diagnostic workflow for suspected gastric tuberculosis. The rationale behind presenting this specific case is to demonstrate how clinicians can systematically navigate diagnostic uncertainty when initial superficial biopsies fail, and the lesion strongly mimics a submucosal tumor.

This representative case is highly appropriate for clinicians and endoscopists seeking practical guidance on managing atypical, refractory gastric lesions that coexist with unexplained systemic findings such as lymphadenopathy¹⁰. By following this multimodal protocol, practitioners can improve early recognition, avoid unnecessary radical resections, and initiate targeted anti-tuberculosis therapy more efficiently in clinical settings¹¹.

Case Presentation:
A 70-year-old woman presented with a one-month history of abdominal pain without typical constitutional symptoms of tuberculosis. Physical examination and baseline laboratory investigations were unremarkable, but gastroscopy revealed an indeterminate subepithelial lesion in the gastric fundus.

Diagnosis, Assessment, and Plan:
Given the non-diagnostic yield of superficial endoscopic biopsies, cross-sectional imaging and endoscopic ultrasound were performed, which revealed concurrent systemic lymphadenopathy and a mass originating from the muscularis propria. To reach a definitive diagnosis, a laparoscopic partial gastrectomy was executed for full-thickness tissue acquisition. Histopathology demonstrated granulomatous inflammation with caseous necrosis, and subsequent tissue polymerase chain reaction (PCR) confirmed the presence of Mycobacterium tuberculosis complex. The patient was successfully managed with standard first-line anti-tuberculosis therapy, resulting in complete symptom resolution without complications.

Protocol

This protocol was performed in compliance with the guidelines of the Institutional Human Research Ethics Committee of Dongguan People’s Hospital under approval number LW 2026-001. Written informed consent was obtained from the patient prior to all diagnostic, therapeutic, and follow-up procedures.

1. Patient eligibility and initial assessment

1. Medical history
   1. Recorded the chief complaint and documented symptom onset, duration, location, character, aggravating factors (e.g., postprandial or nocturnal), and relieving factors.
   2. Recorded associated symptoms (e.g., belching, nausea, vomiting, reflux, bloating, diarrhea, chills, fever, hematochezia/melena) and assessed alarm features (e.g., unintentional weight loss, persistent fever).
   3. Documented past medical history and current medications (e.g., hypertension and type 2 diabetes mellitus; antihypertensive and glucose-lowering therapy).
   4. Screened for tuberculosis-related risk factors (exposure history, prior tuberculosis, immunosuppression) and malignancy-related history.
2. Physical examination
   1. Assessed general status and vital signs (blood pressure, heart rate, temperature, respiratory rate) to confirm clinical stability.
   2. Examined the skin and sclera for jaundice.
   3. Palpated superficial lymph node regions (cervical, supraclavicular, axillary) manually using the pads of the index and middle fingers to assess for enlargement, tenderness, or fixation.
   4. Examined the abdomen systematically via visual inspection, auscultation of bowel sounds, and both light and deep manual palpation to assess for distension, tenderness (including rebound/guarding), palpable masses, and Murphy’s sign.
   5. Examined joints and lower limbs for abnormalities (e.g., edema, erythema, deformity).
3. Initial laboratory investigations
   1. Obtained complete blood count (CBC) using standard laboratory equipment (see Table of Materials).
   2. Obtained liver and renal function tests [alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin, creatinine, urea].
   3. Obtained serum electrolytes [sodium (Na⁺), potassium (K⁺), chloride (Cl⁻)].
   4. Obtained coagulation tests [prothrombin time (PT), activated partial thromboplastin time (APTT)] and cardiac troponin when clinically indicated.
   5. Obtained serum tumor markers [e.g., carcinoembryonic antigen (CEA), carbohydrate antigen 19–9 (CA19–9), alpha-fetoprotein) (AFP)] to support differential diagnosis.
   6. Performed fecal occult blood testing.
   7. Assessed Helicobacter pylori using an institutional method (serology, urea breath test, or stool antigen).
      NOTE: Recorded the method used and its interpretation threshold.

2. Diagnostic workflow for gastrointestinal and systemic lesions

1. Gastroscopy and initial sampling
   1. Performed diagnostic gastroscopy to evaluate the esophagus, stomach (including fundus in retroflex view), and duodenum.
   2. Documented lesion location, size, macroscopic appearance, and suspected layer of origin, and archived representative images.
   3. Classified the duodenal ulcer using a standard endoscopic staging system (e.g., active stage with exudate) and archived images.
   4. Obtained forceps biopsies from the lesion surface and adjacent mucosa (≥6 fragments when technically feasible) using endoscopic forceps (see Table of Materials) and placed specimens in 10% neutral buffered formalin (see Table of Materials).
   5. Labeled specimens with patient identifiers and sampling sites were delivered to pathology within 1 h of collection.
   6. Initiated acid suppression therapy for the active ulcer and reassessed symptom response during follow-up.
2. Cross-sectional imaging
   1. Performed non-contrast chest computed tomography (CT) using a standard adult protocol (e.g., 120 kVp; automatic tube current modulation; pitch 0.9–1.2).
   2. Reconstructed images at 1.0 mm slice thickness with multiplanar reformats and reviewed using lung and mediastinal windows.
   3. Documented pulmonary nodules (location, maximal diameter, density) and mediastinal/hilar lymphadenopathy (stations, short-axis diameter).
   4. Compared with prior imaging when available to assess interval changes.
3. Contrast-enhanced upper abdominal CT
   1. Performed contrast-enhanced upper abdominal CT unless contraindicated.
   2. Administered iodinated contrast (see Table of Materials) at 1.5 mL/kg (max 120 mL) via power injector at 3.0 mL/s, followed by a 30–40 mL saline (see Table of Materials) flush.
   3. Acquired arterial phase at 25–30 s and portal venous phase at 60–70 s after injection, and reconstructed at 1.0–1.25 mm slice thickness.
   4. Documented gastrointestinal wall thickening/enhancement (duodenal bulb, cardia, gastric antrum) and mapped abdominal lymphadenopathy (pericardial, hepatogastric, retroperitoneal) by location and size.
4. Endoscopic ultrasound (EUS) and sampling
   1. Performed EUS using a linear-array echoendoscope (typical frequency range 5–12 MHz; see Table of Materials).
   2. Localized the lesion and determined the layer of origin (e.g., muscularis propria) and recorded lesion size (e.g., 12.4 mm × 5.3 mm), echogenicity, homogeneity, and posterior acoustic features with image archiving.
   3. Performed EUS-guided fine-needle aspiration/biopsy to improve tissue yield for submucosal lesions.
   4. Used a 22-gauge needle (see Table of Materials) and performed 3 passes (2 passes with suction using a 10 mL syringe; 1 pass without suction), with 10–15 to-and-fro movements per pass under continuous ultrasound visualization.
   5. Prepared direct smears from each pass; fixed one slide immediately in 95% ethanol for cytology and air-dried one slide for rapid stain if available.
   6. Placed residual material into 10% neutral buffered formalin to create a cell block for histology and placed a separate portion into a sterile tube for molecular testing.
   7. Monitored for immediate complications (pain, bleeding, perforation) for at least 2 h post-procedure and documented vital signs and symptom status.
      ​NOTE: If the presented case did not include fine-needle aspiration (FNA)/fine-needle biopsy (FNB), replace Steps 2.4.3–2.4.7 with the exact sampling method used and state the limitation explicitly.

3. Surgical intervention and confirmatory laboratory testing

1. Laparoscopic partial gastrectomy
   1. Confirmed the absence of surgical contraindications (e.g., uncontrolled coagulopathy, unstable cardiopulmonary disease) and reviewed preoperative laboratory results.
   2. Induced general anesthesia and established pneumoperitoneum at 12–14 mmHg.
   3. Inserted standard laparoscopic ports (e.g., 10 mm umbilical camera port plus 2–3 working ports; see Table of Materials) and performed systematic exploration.
   4. Localized the lesion based on preoperative endoscopic/imaging information (e.g., fundus near the cardia/greater curvature).
   5. Performed wedge/partial gastrectomy to remove the lesion with a gross margin of ≥1 cm when anatomically feasible and achieved hemostasis and secure closure per institutional practice.
   6. Delivered the specimen to pathology immediately, recorded time-to-fixation, and oriented the specimen (proximal/distal or anterior/posterior) for gross assessment.
2. Histopathology
   1. Fixed tissue in 10% neutral buffered formalin for 24–48 h at room temperature.
   2. Sampled representative areas (lesion center, interface with adjacent tissue, and grossly necrotic areas when present), processed for paraffin embedding, and cut sections at 3–4 µm thickness.
   3. Performed hematoxylin and eosin (H&E) staining and assessed for granulomatous inflammation and necrosis (including caseous necrosis).
   4. Performed special stains: periodic acid-schiff stain (PAS) and periodic acid-silver methenamine stain (PASM) for fungal elements, and Ziehl–Neelsen or auramine–rhodamine staining for acid-fast bacilli.
      ​NOTE: A negative acid-fast bacillus stain (AFB) did not exclude tuberculosis; interpretation was integrated with molecular testing and clinical/imaging findings.
3. Molecular testing [tissue polymerase chain reaction (PCR)]
   1. Selected tissue for molecular testing using clean instruments to avoid cross-contamination and used separate work areas for pre- and post-amplification steps.
   2. Extracted DNA from fresh tissue (preferred) or formalin-fixed, paraffin-embedded (FFPE) tissue using a validated extraction kit. For FFPE tissue, deparaffinized sections with xylene (2 × 10 min) followed by graded ethanol washes (100%, 95%, 70%) (see Table of Materials for reagents), and air-dried before lysis.
   3. Performed real-time PCR targeting M. tuberculosis complex (e.g., IS6110 or another validated target).
   4. Used the following cycling conditions as a standard real-time PCR profile: initial denaturation at 95 °C for 5 min; 40 cycles of 95 °C for 15 s and 60 °C for 60 s.
   5. Interpreted results using laboratory thresholds [e.g., cycle threshold (Ct) ≤ 38 as positive, Ct 38–40 as indeterminate requiring repeat testing, and no amplification as negative] and confirmed indeterminate results by repeat extraction and repeat PCR.
      NOTE: Replace target gene(s) and Ct thresholds with those used by the laboratory if different.
4. Interferon-gamma release assay (IGRA)
   1. Performed whole-blood IGRA to support tuberculosis infection assessment.
   2. Collected 1 mL blood into each assay tube [nil control, tuberculosis bacillus (TB) antigen tubes, mitogen control] and mixed by inverting 10 times immediately after collection.
   3. Incubated tubes at 37 °C for 16–24 h, centrifuged at 2,000–3,000 × g for 15 min, and harvested plasma for enzyme-linked immunosorbent assay (ELISA) measurement of interferon-gamma.
   4. Interpreted IGRA as positive when (TB antigen − nil) ≥0.35 IU/mL and ≥25% of nil, with an adequate mitogen response per assay criteria.
      ​NOTE: If IGRA was not performed in this case, state this explicitly and provide the reason.
5. Diagnostic integration
   1. Established the diagnosis by integrating (i) granulomatous inflammation with caseous necrosis on histopathology, (ii) molecular evidence of M. tuberculosis complex on PCR, and (iii) supportive clinical/imaging findings.
   2. Recorded alternative diagnoses considered (e.g., malignancy, lymphoma, fungal infection) and the negative findings supporting exclusion.

4. Therapeutic management and follow-up-anti-tuberculosis therapy

1. Consulted infectious diseases specialists to confirm the regimen after diagnostic confirmation.
2. Administered standard first-line anti-tuberculosis therapy (rifampicin, isoniazid, pyrazinamide, ethambutol) per institutional/national guidance.
3. Adjusted doses based on body weight and hepatic/renal function and documented baseline and follow-up laboratory values.
4. Monitored for adverse events (gastrointestinal intolerance, hepatotoxicity, rash, visual symptoms) and managed them per institutional practice.
5. Scheduled follow-up at 2 weeks after discharge and then every 4 weeks during early treatment, with subsequent intervals individualized based on clinical response.
6. Performed CBC and liver function tests at each visit.
7. Evaluated symptom resolution and screened for recurrence based on clinical assessment and repeat imaging/endoscopy when clinically indicated.

Results

Initial assessment results
A 70-year-old woman presented with abdominal pain for one month. She reported no known tuberculosis exposure and denied constitutional or alarm symptoms, including nausea, vomiting, fever, and significant weight loss. Her medical history included well-controlled hypertension and type 2 diabetes mellitus, managed with chronic oral medications. On physical examination, she was alert and hemodynamically stable. No jaundice or palpable superficial lymphadenopathy was noted. The...

Discussion

Gastric tuberculosis is an uncommon manifestation of gastrointestinal tuberculosis and can arise through hematogenous or lymphatic dissemination, direct extension from adjacent lesions, or implantation in the setting of impaired mucosal defense. Because symptoms are non-specific and endoscopic appearances often overlap with peptic ulcer disease, malignancy, lymphoma, and subepithelial tumors, diagnosis is frequently delayed and may require a stepwise approach that integrates clinical context with tissue-based confirmatio...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Acid-fast stain kit (Ziehl–Neelsen)	BD Difco	215073	For AFB staining; auramine stain acceptable alternative
Centrifuge (2,000–3,000 × g)	Eppendorf	5810 R	Plasma separation for IGRA
Coverslips	Citotest	10212424C	For microscopy preparation
CT scanner (multi-detector)	Siemens Healthineers	SOMATOM Definition AS+	Any ≥64-slice CT acceptable; record kVp, slice thickness, phases
Disposable biopsy forceps	Boston Scientific	1395 (Radial Jaw 4)	For mucosal/surface biopsies; any sterile single-use forceps acceptable
DNA extraction kit (FFPE)	QIAGEN	56404 (QIAamp DNA FFPE Tissue Kit)	If using FFPE sections for PCR
DNA extraction kit (tissue)	QIAGEN	69504 (DNeasy Blood & Tissue Kit)	For fresh tissue DNA extraction
ELISA plate reader	BioTek (Agilent)	Synergy H1	For IGRA ELISA readout (if using kit requiring reader)
EUS ultrasound processor	Olympus	EU-ME2	Any equivalent EUS processor acceptable
EUS-FNA/FNB needle, 22G	Cook Medical	EchoTip Ultra (22G)	If your case used FNA/FNB; record gauge and number of passes
Gastroscope	Olympus	GIF-HQ190	Standard diagnostic gastroscope; equivalent models acceptable
Grocott’s methenamine silver (GMS/PASM) kit	Abcam	ab150680	For fungal screening (PASM/GMS)
H&E staining kit or reagents	Sigma-Aldrich (Merck)	HT110116	Hematoxylin and eosin staining
IGRA test kit	Qiagen	QuantiFERON-TB Gold Plus	Whole-blood IGRA; record interpretation criteria
Image analysis / labeling software	NIH	ImageJ/Fiji	For adding scale bars/annotations to histology images
Insufflator	Karl Storz	ENDOMAT	Maintain pneumoperitoneum; record pressure used
Laparoscope	Karl Storz	26003BA	10 mm laparoscope; equivalent acceptable
Laparoscopic specimen retrieval bag	Applied Medical	Endo Catch II	For safe extraction of specimen
Laparoscopic stapler (wedge/partial gastrectomy)	Medtronic	Endo GIA	Use institutional stapler; record cartridge type if relevant
Laparoscopic tower (camera + light source)	Karl Storz	IMAGE1 S	Any equivalent laparoscopic imaging system acceptable
Linear EUS echoendoscope	Olympus	GF-UCT180	Linear EUS recommended for sampling; equivalent acceptable
Microscope slides	Citotest	188105	For cytology smears
Microtome	Leica Biosystems	RM2235	For 3–4 μm sectioning
Mounting medium	Sigma-Aldrich (Merck)	6522	Coverslipping
MTB complex PCR kit (validated)	Sansure Biotech	MTB DNA Diagnostic Kit	Use your institution’s validated MTB complex PCR; record target and Ct threshold
Nonionic iodinated contrast (for CT)	GE Healthcare	Omnipaque 350	Use institutional standard contrast; record concentration and dose
Paraffin embedding system	Leica Biosystems	HistoCore Arcadia	Any equivalent embedding system acceptable
PAS staining kit	Sigma-Aldrich (Merck)	395B	For fungal screening
PCR clean workflow supplies (filter tips)	Thermo Fisher Scientific	2069G	Use aerosol-resistant tips to reduce contamination
Power injector for contrast	Bayer	MEDRAD Stellant	Any injector acceptable; record injection rate and saline flush
Real-time PCR instrument	Roche	LightCycler 480 II	Any validated real-time PCR system acceptable
Specimen container (leak-proof)	Thermo Fisher Scientific	02-544-208	For formalin-fixed tissue transport
Statistical/figure software	GraphPad Software	Prism	If used for plotting; optional
Syringe (10 mL)	BD	309604	For suction during FNA when used
Video gastroscope system (diagnostic upper GI endoscopy)	Olympus	CV-190 (processor), CLV-190 (light source)	Any equivalent endoscopy platform is acceptable; record model used
10% Neutral Buffered Formalin	Sigma-Aldrich (Merck)	HT501128	Fixation of biopsy and resection specimens
37 °C incubator	Thermo Fisher Scientific	Heratherm	For IGRA incubation
95% Ethanol	Sigma-Aldrich (Merck)	459844	Immediate fixation of cytology slides