Non-traumatic necrotizing fasciitis caused by Bacillus cereus in a child with acute lymphoblastic leukemia: a case report and proposed gastrointestinal-to-hematogenous pathway Authors: Yue Liu, Xianzheng Lin, Jian Zhang, Tao Tang, Yueling Xi, Lingling Zheng, Sijuan Sun, Wenlan Zhang Published: 2026 ABSTRACT AbstractBackgroundBacillus cereus (B. cereus) is generally considered a self-limiting foodborne pathogen but can cause severe invasive infections in immunocompromised hosts. Necrotizing fasciitis (NF) due to B. cereus is exceedingly rare, particularly in pediatric oncology patients. Moreover, hematogenous NF without antecedent trauma is uncommon, and its pathogenesis remains poorly understood. This case is reported to highlight a potential gastrointestinal-to-hematogenous route of infection and its clinical implications.Case presentationA 10-year-old boy with acute lymphoblastic leukemia developed fulminant, non-traumatic NF and septic shock during chemotherapy-induced myelosuppression. The illness began with gastrointestinal symptoms after food intake and rapidly progressed to extensive soft tissue necrosis and circulatory collapse. B. cereus was isolated from both blood and tissue cultures. Due to profound neutropenia, severe thrombocytopenia, coagulopathy, and hemodynamic instability, immediate radical surgical debridement was considered prohibitively risky. The patient was initially treated with broad-spectrum antimicrobial therapy, intensive organ support, and correction of coagulopathy. Following clinical stabilization, staged surgical debridement and vacuum-assisted closure were performed. The patient recovered and was able to resume leukemia treatment.ConclusionThis case expands the clinical spectrum of invasive B. cereus infection and suggests a potential paradigm in which gastrointestinal translocation may lead to hematogenous NF in immunocompromised hosts, even in the absence of trauma. It underscores the importance of early recognition, multidisciplinary collaboration, and individualized decision-making regarding surgical timing. Further studies are warranted to better define the pathogenesis and optimal management strategies for this rare but life-threatening condition.Keywords: Bacillus cereus, Necrotizing fasciitis, Acute lymphoblastic leukemia, Immunocompromised host Introduction Acute lymphoblastic leukemia (ALL) is the most common malignancy in children and is frequently complicated by severe infections due to chemotherapy-induced immunosuppression. Among these, bacterial infections remain a leading cause of morbidity and mortality, particularly during periods of profound neutropenia [1, 2]. NF is a rapidly progressive and life-threatening soft tissue infection characterized by widespread fascial necrosis and systemic toxicity. It is most commonly caused by Group A Streptococcus or polymicrobial pathogens and is typically associated with disruption of the skin barrier [3–5]. In contrast, hematogenous dissemination leading to NF is uncommon, especially in pediatric patients, making early recognition particularly challenging in the absence of an obvious portal of entry. B. cereus is a ubiquitous, spore-forming Gram-positive bacterium widely present in the environment. Although it is best known as a cause of self-limiting foodborne gastroenteritis, increasing evidence indicates that it can act as an opportunistic pathogen in immunocompromised hosts, leading to severe invasive infections such as bacteremia, central nervous system infection, and sepsis [6, 7]. However, NF caused by B. cereus is exceedingly rare, particularly in children with hematologic malignancies, and its clinical characteristics, pathogenesis, and optimal management remain poorly defined. Here, we report a case of fulminant, non-traumatic NF and septic shock caused by B. cereus in a child with ALL during chemotherapy-induced myelosuppression. Notably, the clinical course suggests a potential gastrointestinal-to-hematogenous route of infection in the absence of trauma or skin disruption. This case highlights the importance of early recognition of atypical presentations and supports an individualized, multidisciplinary management strategy in this high-risk population. Case report A 10-year-old boy was recently diagnosed with high-risk B-cell acute lymphoblastic leukemia (ALL) and received induction chemotherapy. Due to poor early treatment response (day 19 minimal residual disease [MRD] was 63.79%), he enrolled in a clinical trial incorporating trametinib and sirolimus. Subsequently, he experienced severe chemotherapy-related myelosuppression. During the subsequent course, the patient developed fulminant B. cereus septic shock and NF during profound neutropenia. The chronological clinical course is summarized in Table 1. Table 1Timeline of disease progression, key investigations, and therapeutic interventionsTimeClinical courseKey findingsManagement and clinical decision-makingDay 0Diagnosis of high-risk B-ALLBone marrow examination confirmed B-ALLInduction chemotherapy with the PVDL regimenDay 19Poor early response to induction therapyMRD 63.79%Enrollment in a clinical trial of trametinib plus sirolimusDay 25Recurrent vomiting with small-volume hematemesis and persistent epigastric pain after consuming takeout food; severe right lower-extremity pain developed 10 h later, followed by rapidly progressive erythematous and violaceous bullous skin lesions involving both lower extremitiesWBC 0.27 × 109 /L; ANC 0.07 × 109 /L; PLT 47 × 109 /L; Hb 74 g/L; CRP 10.8 mg/L; PCT 1.11 μg/L. Abdominal CT demonstrated diffuse gastric wall edemaEmpirical levofloxacin; tramadol; platelet transfusion; bowel rest; intravenous fluids; acid suppression and antiemetic therapyEarly Day 26Progressive circulatory failure and shockWBC 0.11 × 109 /L; ANC 0.04 × 109 /L; PLT 92 × 109 /L; Hb 64 g/L; CRP 14.5 mg/L; PCT 3.54 μg/L; lactate 8.6 mmol/LPICU admission; meropenem, vancomycin and caspofungin; fluid resuscitation; red blood cell transfusion; vasoactive support (peak VIS 30)Days 26–27Rapid progression of skin lesions with epidermal detachment, hemorrhagic exudation, and worsening bilateral lower-extremity painWBC 0.22 × 109 /L; ANC 0.07 × 109 /L; PLT 16 × 109 /L; CRP 65.4 mg/L; PCT 31.75 μg/L; IL-6 2366.09 pg/mL; IL-8 7810.21 pg/mL; IL-1β 31.23 pg/mL; PT 19.9s; APTT 60.6s; Fib 1.18 g/L; thrombomodulin 18.5 TU/mL; TAT complex 6.1 ng/mL. Blood mNGS positive for Bacillus cereus. Ultrasound of the lower extremities showed subcutaneous tissue thickening with fissure-like hypoechoic areasMDT consultation. Septic shock, sepsis-associated coagulopathy, cytokine storm, and NF were suspected. TPE, CRRT, methylprednisolone, and tocilizumab were initiated. Emergency radical debridement was deferred because of profound neutropenia, thrombocytopenia, coagulopathy, and hemodynamic instabilityDays 28–30Hemodynamic status gradually improved; skin lesions stabilizedBlood and wound cultures positive for Bacillus cereus (MALDI-TOF MS). Antimicrobial susceptibility testing demonstrated susceptibility to vancomycin, meropenem, and clindamycin and resistance to penicillin G. MRI showed extensive soft tissue edema without definitive bone involvementVasoactive agents discontinued on PICU day 3; CRRT discontinued after 4 days; Continued organ support and local wound care; vancomycin dose adjusted according to therapeutic drug monitoringDays 31–39Recovery of hemodynamic and organ function; persistent bone marrow suppressionBlood cultures became negative on Day 39. WBC 1.23 × 109 /L; ANC 0.96 × 109 /L; PLT 34 × 109 /L; Hb 60 g/L; CRP 2.1 mg/L; PCT 0.12 μg/Lmeropenem completed after 14 days; continued vancomycin and local wound management; transfuse packed red blood cells, platelets and fresh frozen plasma as needed.Day 40First damage-control debridementANC 0.62 × 109 /L; PLT 69 × 109 /L. Extensive necrosis involving skin, subcutaneous fat, fascia, and muscle with perforator vessel thrombosis. Histopathology confirmed NFStaged surgical debridementDays 41–61Serial debridement and wound reconstructionProgressive demarcation between viable and nonviable tissue; infection remained controlledMultiple debridement, VAC therapy, skin grafting, and flap reconstruction; vancomycin continued for 6 weeksDay 50Completion of PICU treatmentStable vital signs and recovery of organ functionTransfer to the hematology ward after 25 days in PICUDay 62Marked clinical improvementBone marrow MRD 1.05%Resumption of anti-leukemia therapy1-year follow-upComplete wound healing with residual scarring; normal lower-extremity function; no recurrent infectionMRD negativeContinued leukemia treatment and follow-upAbbreviations: ALL, acute lymphoblastic leukemia; MRD, minimal residual disease; WBC, white blood cell count; ANC, absolute neutrophil count; PLT, platelet count; Hb, hemoglobin; CRP, C-reactive protein; PCT, procalcitonin; CT, computed tomography; PICU, pediatric intensive care unit; VIS, vasoactive-inotropic score; IL, interleukin; PT, prothrombin time; APTT, activated partial thromboplastin time; Fib, fibrinogen; TAT, thrombin–antithrombin; mNGS, metagenomic next-generation sequencing; MDT, multidisciplinary team; NF, necrotizing fasciitis; TPE, therapeutic plasma exchange; CRRT, continuous renal replacement therapy; MRI, magnetic resonance imaging; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; VAC, vacuum-assisted closure Timeline of disease progression, key investigations, and therapeutic interventions Abbreviations: ALL, acute lymphoblastic leukemia; MRD, minimal residual disease; WBC, white blood cell count; ANC, absolute neutrophil count; PLT, platelet count; Hb, hemoglobin; CRP, C-reactive protein; PCT, procalcitonin; CT, computed tomography; PICU, pediatric intensive care unit; VIS, vasoactive-inotropic score; IL, interleukin; PT, prothrombin time; APTT, activated partial thromboplastin time; Fib, fibrinogen; TAT, thrombin–antithrombin; mNGS, metagenomic next-generation sequencing; MDT, multidisciplinary team; NF, necrotizing fasciitis; TPE, therapeutic plasma exchange; CRRT, continuous renal replacement therapy; MRI, magnetic resonance imaging; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; VAC, vacuum-assisted closure On day 25 of treatment, approximately 6 hours after consuming takeout food, he developed five episodes of vomiting, with a small amount of fresh blood in the vomitus, accompanied by persistent upper abdominal pain without fever. Abdominal computed tomography (CT)scan showed diffuse edema of the gastric wall (Fig. 1). About 10 hours after vomiting onset, the patient suddenly experienced severe right lower extremity pain; at that time, the overlying skin appeared normal. Two hours later, a 3 cm × 3 cm erythematous patch developed in the right inguinal region, with marked tenderness, no fluctuation, no crepitus, and mild local warmth. The lesion gradually became raised, expanded in size, and rapidly progressed into an 8 cm × 8 cm dark purple bullous plaque. Within a few hours, similar lesions appeared on both lower extremities, accompanied by severe pain (Fig. 2). No history of trauma, surgery, or skin breach was reported. Notably, the onset of pain preceded visible skin changes and was disproportionate to the initial cutaneous findings. Laboratory tests at that time showed a white blood cell (WBC) count of 0.27 × 109 /L, absolute neutrophil count (ANC) 0.07 × 109 /L, platelet count (PLT) 47 × 109 /L, C-reactive protein (CRP) 10.8 mg/L, and procalcitonin (PCT) 1.11 μg/L. These results indicated severe myelosuppression. Despite profound neutropenia, the patient remained afebrile early in the disease course and therefore did not meet the diagnostic criteria for febrile neutropenia. The patient was not receiving systemic antibacterial therapy immediately before symptom onset. Given the initial clinical presentation, a gastrointestinal infection or foodborne illness was considered. Although the etiology of the cutaneous lesions was initially unclear, empiric levofloxacin was administered, alongside tramadol for analgesia, nil per os, intravenous fluids, acid suppression, antiemetics, and platelet transfusion; however, the condition rapidly progressed. Abdominal CT findings after onset of gastrointestinal symptoms. Axial contrast-enhanced CT image showing diffuse gastric wall thickening and edema (black arrows) Rapid progression of necrotizing soft-tissue lesions. A large violaceous-to-black necrotic skin lesion with surrounding erythema and edema developed on the right thigh near the knee. Dotted lines indicate the clinical margins of progression In the early hours of day 26, the patient developed persistent tachycardia, hypotension, elevated lactate (8.6 mmol/L), and tissue hypoperfusion, progressing to shock, and was urgently transferred to the pediatric intensive care unit (PICU). At that time, WBC was 0.11 × 109 /L, ANC 0.04 × 109 /L, CRP 14.5 mg/L, and PCT 3.54 μg/L. Septic shock was the primary clinical consideration. Empiric antimicrobial therapy was escalated to meropenem, vancomycin, and caspofungin. Red blood cell transfusions, fluid resuscitation, and vasoactive agents were administered, with a peak vasoactive-inotropic score (VIS) of 30. Nevertheless, cutaneous lesions further progressed, with partial epidermal detachment, exudation, and surrounding dark red ischemic necrosis. On the afternoon of day 26, laboratory tests revealed a marked increase in inflammatory markers: CRP 65.4 mg/L, PCT 31.75 μg/L, interleukin-6 (IL-6)2366.09 pg/mL, alongside severe coagulopathy (prothrombin time(PT)19.9 s, activated partial thromboplastin time(APTT) 60.6 s, fibrinogen 1.18 g/L) and endothelial injury with abnormal thrombomodulin. Ultrasound of the lower extremities showed subcutaneous tissue thickening with fissure-like hypoechoic areas. Blood metagenomic next-generation sequencing (mNGS) the following day detected B. cereus. In response, a multidisciplinary team (MDT) involving PICU, hematology, infectious disease, and reconstructive surgery was convened. The patient was considered to have septic shock with sepsis-induced coagulopathy, cytokine storm, and suspected NF. However, given profound neutropenia, severe thrombocytopenia, coagulopathy, and persistent hemodynamic instability (VIS 30), immediate radical debridement posed a high risk of catastrophic hemorrhage and perioperative mortality. The MDT therefore opted for an individualized staged strategy: first, aggressive medical management to stabilize systemic status, followed by delayed surgical intervention. On day 30, blood and wound cultures confirmed B. cereus. Species identification was performed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), and antimicrobial susceptibility testing was performed using the VITEK 2 automated system, showing sensitivity to vancomycin, meropenem, and clindamycin, and resistance to penicillin G. No further virulence gene or toxin typing was conducted. Concurrent MRI of both lower extremities demonstrated extensive soft tissue edema without definitive bone involvement. During treatment, the patient received meropenem (14 days), vancomycin (6 weeks, target trough 10–20 mg/L), and caspofungin. Given ongoing severe neutropenia and high risk for secondary infections, meropenem was continued until neutrophil recovery despite available susceptibility results. Supportive care included fluid resuscitation, vasoactive support, therapeutic plasma exchange (TPE) for coagulopathy, continuous renal replacement therapy (CRRT) for cytokine clearance, and immunomodulation with methylprednisolone and tocilizumab. Local wounds were disinfected with povidone-iodine and covered with sterile dressings. With intensive therapy, circulatory function improved, inflammatory markers decreased, vasoactive drugs were discontinued after 3 days, and CRRT was stopped after 4 days. Bone marrow suppression persisted, and PLT remained < 50 despite transfusions, presenting high operative bleeding risk. As infection gradually came under control and hematopoietic function recovered, inflammatory markers continued to decline. On day 39, blood cultures turned negative. On day 40, ANC recovered to 0.62 × 109 /L, PLT to 69 × 109 /L, and coagulation improved, allowing the first damage-control debridement. Intraoperatively, multiple necrotic lesions were observed on both lower extremities, with extensive involvement of skin, subcutaneous fat, deep fascia, and some muscle, along with perforator vessel thrombosis. Necrotic tissue was removed, and vacuum-assisted closure (VAC) was applied (Figs. 3A–C). Surgical management of necrotizing fasciitis. A. Intraoperative findings before the first debridement. Multiple necrotic skin lesions were present on the right lower extremity, including black–purple necrotic plaques on the right thigh near the knee and the right lower leg. A yellow–white area of moist gangrene was observed in the right inguinal region. Black arrows indicate the locations from which tissue samples were collected for pathological examination. B. Wound appearance immediately after the first debridement. Extensive necrotic skin, subcutaneous tissue, fascia, and nonviable soft tissue were removed, resulting in multiple full-thickness soft-tissue defects. C. Application of vacuum-assisted closure (VAC) therapy following surgical debridement. Negative-pressure wound therapy was used to promote wound drainage, control local infection, and facilitate granulation tissue formation Histopathology revealed that lesions primarily involved deep fascia and subcutaneous fat, with extensive structural destruction and necrosis. Involved fascia and fat contained intravascular thrombi tightly adherent to the damaged endothelium, along with fibrinoid necrosis of the vessel wall, consistent with in situ thrombus formation rather than postmortem clotting, reflecting acute ischemic injury. Extensive microvascular occlusion caused ischemic necrosis of fascia, fat, and overlying skin. Notably, necrotic fascia and fat exhibited minimal inflammatory cell infiltration, suggesting limited host immune response under profound immunosuppression. In contrast, inflammatory cells were present at the interface between necrotic and viable tissue, with nuclear debris formation. Epidermal structure at lesion margins appeared relatively preserved at low magnification, dermis showed edema, and inflammatory cells extended into deep fat and fascia (Figs. 4A–F). This demonstrates a characteristic “deep-to-superficial” invasion pattern of NF, where deep fascia and subcutaneous fat are most severely affected, and epidermal necrosis is secondary and delayed. This histopathologic pattern correlates with clinical findings, explaining why early skin signs may be subtle despite extensive deep tissue necrosis and vascular occlusion. Histopathological findings of necrotic tissue in different regions of the right lower limb (H&E stain). Right thigh near the knee: A (HE × 1): deep fascia and subcutaneous adipose tissue showing extensive degeneration and necrosis, interstitial vascular dilation, congestion, and thrombosis; adipose tissue architecture disrupted, with erythrocyte extravasation. B (HE × 40): vascular wall structures are blurred with fibrinoid necrosis (black arrows); nuclei are pyknotic or absent, smooth muscle layer disrupted; intraluminal thrombus tightly adherent to the damaged endothelium. C (HE × 1): margin of completely necrotic skin on the anteromedial region above the knee, with most of the epidermis absent and covered by eschar; dermis and subcutaneous adipose tissue show diffuse necrosis, with inflammatory cell infiltration at the lesion edge (black arrows) and marked interstitial edema. D (HE × 40): In the margin region, mononuclear inflammatory cells infiltrate between collagen fibers of the dermis and subcutaneous adipose tissue (black arrows), with adjacent erythrocyte extravasation and nuclear debris from dead inflammatory cells. Medial proximal thigh: E (HE × 1): margin of necrotic skin and fascia, with epidermis covered by fibrinous exudate and necrotic crust; dermis shows edema and inflammatory cell infiltration, and erythrocyte extravasation at the lesion edge (black arrows); underlying subcutaneous adipose septa widened with structural disruption. F (HE × 40): deep dermis and subcutaneous adipose tissue at the necrotic margin show diffuse inflammatory cell infiltration (black arrows), with focal microabscess formation Subsequently, the patient underwent multiple staged debridement, VAC therapy, flap reconstruction, and skin grafting. After comprehensive treatment, vital signs stabilized and organ function recovered. The patient was transferred from PICU to the hematology ward after 25 days. On day 62, bone marrow MRD decreased to 1.05%, and anti-leukemia therapy was resumed. Continuous monitoring showed rapid elevation of inflammatory markers during septic shock, followed by gradual decline with anti-infective and organ-supportive therapy. Recovery of ANC and PLT paralleled clinical improvement and ultimately enabled delayed surgical debridement (Fig. 5). During 1-year follow-up, no further invasive infections occurred, wounds fully healed with residual scarring, muscle strength of bilateral lower extremities was normal, and the last bone marrow evaluation confirmed MRD negativity. Temporal trends of inflammatory markers and hematologic parameters during the clinical course. A. Serial variations of C-reactive protein (CRP), procalcitonin (PCT), and interleukin-6 (IL-6). Key clinical events are indicated, including PICU admission, initiation of continuous renal replacement therapy (CRRT) and therapeutic plasma exchange (TPE), administration of methylprednisolone and tocilizumab, blood culture negative, and first damage-control debridement. B. Dynamic changes in absolute neutrophil count (ANC) and platelet count (PLT). Major clinical events, including PICU admission, blood culture negative, and first damage-control debridement are indicated Discussion In this patient, B. cereus caused fulminant NF and septic shock in the setting of profound immunosuppression. Although traditionally regarded as a relatively benign foodborne pathogen, accumulating evidence indicates that B. cereus can cause severe invasive disease in immunocompromised hosts, particularly in patients with hematologic malignancies and prolonged neutropenia [8–10]. However, NF attributable to B. cereus remains exceedingly rare, especially in pediatric oncology patients, and its pathogenesis and optimal management have not been fully elucidated. A notable feature of this case is the absence of antecedent trauma or skin disruption, which are typically considered prerequisites for NF. Instead, gastrointestinal symptoms preceded the onset of bacteremia and rapidly progressive soft tissue infection. In the absence of an alternative portal of entry, this temporal sequence supports a gastrointestinal-to-hematogenous route of infection. We propose a three-step pathogenic model: (1) chemotherapy-induced mucosal barrier injury facilitating intestinal translocation of B. cereus; (2) hematogenous dissemination in the context of profound neutropenia; and (3) toxin-mediated vascular endothelial injury and sepsis-associated coagulopathy, resulting in microvascular thrombosis, tissue ischemia, and extensive fascial necrosis. This proposed mechanism is consistent with the known virulence profile of B. cereus, including its ability to produce cytotoxins and induce endothelial damage [11]. Histopathological findings in this case provide further support for this mechanism. Extensive intravascular thrombosis was observed within the fascia and adipose tissue, with thrombi adherent to injured vascular endothelium, indicating in situ thrombosis and acute ischemic injury. The affected tissues showed widespread necrosis with minimal inflammatory cell infiltration, likely reflecting both rapid toxin-mediated tissue destruction and impaired host immune response under severe immunosuppression. Inflammatory cells were mainly concentrated at the interface between necrotic and viable tissue. These findings further confirm that the primary pathological process of NF originates in the deep fascia and subcutaneous tissue, whereas skin involvement is largely secondary. They also explain why early cutaneous manifestations may underestimate the true extent of underlying disease. Although clindamycin is well recognized for its ability to suppress bacterial toxin production in toxin-mediated soft tissue infections [12], it was not administered in this case. Following escalation to meropenem and vancomycin-based therapy, the patient demonstrated progressive clinical improvement with stabilization of hemodynamics and resolution of inflammatory markers. Although subsequent susceptibility testing showed that the isolate was susceptible to clindamycin, additional antimicrobial modification was deemed unnecessary given the favorable clinical trajectory. Nonetheless, its potential role in mitigating toxin-mediated tissue injury should be acknowledged and may be considered in similar future cases if initial response to broad-spectrum therapy is suboptimal. A review of the literature indicates that invasive B. cereus infection is uncommon but associated with high mortality in children with hematologic malignancies. Reported manifestations include fulminant sepsis, central nervous system infection, pneumonia, liver abscesses, and catheter-related bloodstream infections, most commonly occurring during periods of profound chemotherapy-induced neutropenia. In contrast, NF caused by B. cereus has only been reported sporadically in this population. Most previously reported cases were characterized by rapidly progressive soft-tissue infection accompanied by severe sepsis or septic shock and were associated with poor outcomes [13, 14]. Consistent with previous reports, profound neutropenia was a shared risk factor. However, unlike most published cases, our patient had no history of skin disruption or invasive procedures. The occurrence of gastrointestinal symptoms preceding bacteremia and soft-tissue necrosis further supports gastrointestinal translocation followed by hematogenous dissemination and expands the clinical spectrum of invasive B. cereus infection in immunocompromised hosts. Moreover, the pathological demonstration of vascular thrombosis and ischemic tissue injury provides additional evidence supporting a toxin-mediated mechanism of endothelial damage. This case also raises an intriguing observation. Despite clear evidence of bacteremia, tissue injury remained largely confined to the skin, subcutaneous tissue, and fascia, without involvement of bone, joints, or other solid organs. The extensive perforator vessel thrombosis identified histologically suggests that local microvascular dysfunction may play a central role in tissue necrosis. We speculate that the dense and terminal microvascular network of superficial soft tissues may be more susceptible than bone or deep organs to toxin-mediated endothelial injury and sepsis-associated coagulation abnormalities. In addition, prompt initiation of targeted antimicrobial therapy, organ support, and immunomodulatory treatment after bacteremia was identified may have limited further dissemination and prevented complications such as osteomyelitis or visceral abscess formation. Nevertheless, the tissue tropism and organ selectivity of invasive B. cereus infection remain poorly understood, and the underlying mechanisms require further investigation. From a clinical perspective, this case highlights that hematogenous NF should be considered in immunocompromised patients presenting with gastrointestinal symptoms followed by rapidly progressive soft tissue pain, even in the absence of trauma. Notably, our patient exhibited the classic feature of “pain out of proportion” at an early stage, when cutaneous findings remained relatively subtle. This clinical sign may serve as an important warning indicator of deep fascial necrosis. Early aggressive surgical debridement is widely regarded as the cornerstone of NF management [15]. However, outcomes remain poor in patients with hematologic malignancies and profound neutropenia. Importantly, evidence supporting early debridement is derived primarily from studies of the general NF population, whereas data specific to patients with hematologic malignancies are limited. Previous reports have shown that even with timely surgical debridement and appropriate antimicrobial therapy, some severely immunocompromised children still succumbed to persistent neutropenia, uncontrolled infection, or multiple organ failure [16, 17]. These observations suggest that, in this highly vulnerable population, host-related factors—including profound neutropenia, severe thrombocytopenia, and impaired immune function—may be as important as the timing of source control in determining clinical outcomes. Therefore, although early debridement remains the accepted standard of care, its potential benefits must be weighed against perioperative risks in patients with severe coagulopathy and hemodynamic instability. At the peak of illness, our patient had refractory septic shock, profound neutropenia (ANC < 0.1 × 109 /L), severe thrombocytopenia, and significant coagulation abnormalities, creating an extremely high-risk surgical environment. Consequently, the multidisciplinary team elected to defer radical debridement as an individualized risk–benefit decision. Staged surgical debridement was performed after circulatory stabilization, hematologic recovery, and clearer demarcation of necrotic tissue had been achieved, ultimately resulting in effective infection control and preservation of limb function. It should be emphasized that this case does not challenge the established principle of early debridement. Rather, it suggests that, for selected ultra-high-risk patients with temporary contraindications to surgery, an individualized “stabilization-first, surgery-later” strategy under close monitoring may serve as a critical bridge to definitive source control. Further studies are needed to identify which patients require immediate surgery and which may benefit from a carefully monitored delayed-debridement approach. From a public health perspective, this case also raises concerns regarding food safety in immunocompromised pediatric populations. Current dietary recommendations for neutropenic patients may underestimate the risks posed by toxigenic, spore-forming organisms such as B. cereus, which can survive standard food processing and storage conditions [18, 19]. Clinicians should maintain a high index of suspicion for invasive B. cereus infection when immunocompromised patients develop gastrointestinal symptoms followed by rapid systemic deterioration. Limitations. Despite the unique clinical and pathological findings, this report has several limitations. First, this is a single-case report, and therefore the proposed gastrointestinal-to-hematogenous route of infection and the favorable outcome associated with delayed staged debridement cannot be generalized to other patients. Second, although the temporal sequence of gastrointestinal symptoms, bacteremia, and soft-tissue infection strongly suggests gastrointestinal translocation as the source of infection, direct microbiological evidence from stool samples, food specimens, or gastrointestinal tissue was not available. Consequently, the presumed route of infection remains speculative. Third, toxin profiling and virulence gene analysis of the isolated B. cereus strain were not performed. Therefore, the specific virulence factors responsible for endothelial injury, thrombosis, and tissue necrosis could not be determined. Fourth, retrospective Gram staining of archived pathological samples was initiated but unfortunately remained uncompleted within the revision timeframe. Consequently, direct in-situ bacterial evidence and corresponding Gram-stained micrographs could not be integrated into the current manuscript. Nevertheless, the definitive microbiological diagnosis was established through concordant blood cultures, wound tissue cultures, and mNGS. Finally, although delayed surgical intervention was successful in this patient, this approach should not be interpreted as a challenge to the established principle of early surgical debridement for NF. Rather, it reflects an individualized risk–benefit assessment in the setting of profound neutropenia, severe thrombocytopenia, coagulopathy, and refractory septic shock. Further studies are required to identify which patients may benefit from such a staged management strategy. In conclusion, this case expands the clinical spectrum of invasive B. cereus infection and suggests a potential paradigm in which gastrointestinal translocation may lead to hematogenous NF in immunocompromised hosts, even in the absence of trauma. It underscores the importance of early recognition, multidisciplinary collaboration, and individualized decision-making regarding surgical timing. Further studies are warranted to better define the pathogenesis and optimal management strategies for this rare but life-threatening condition. Acknowledgements The authors would like to express their sincere gratitude to the patient for providing consent to publish this case report. We also extend our appreciation to the medical staff involved in the patient’s care. Authors’ contributions Liu Y and Lin XZ contributed equally to this work. Zhang J and Tang T share senior authorship as co-corresponding authors and jointly supervised this study. Liu Y and Lin XZ were responsible for study conception and design, data collection, and drafting of the manuscript. Xi YL, Zheng LL, Sun SJ and Zhang WL contributed to patient management, clinical data collection, and interpretation of data. Zhang J and Tang T provided critical revision of the manuscript for important intellectual content and overall supervision of the study. All authors read and approved the final manuscript and agree to be accountable for all aspects of the work. Funding This study received no external funding. Data availability The datasets used and analyzed during the current study are not publicly available due to patient privacy restrictions but are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate This case report was approved by the Ethics Committee of Shanghai Children’s Medical Center. The study was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from the patient’s legal guardians. Consent for publication Written informed consent for publication was obtained from the patient’s legal guardians. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yue Liu, Xianzheng Lin, Jian Zhang and Tao Tang contributed equally to this work. References Associated Data Data Availability Statement The datasets used and analyzed during the current study are not publicly available due to patient privacy restrictions but are available from the corresponding author on reasonable request.