Spondylodiscitis caused by Citrobacter koseri: a case report

Abstract

Background

Citrobacter species are ubiquitous environmental gram-negative bacilli that rarely cause severe infections in immunocompetent adults. Spondylodiscitis caused by Citrobacter koseri is an exceptionally rare clinical entity, with fewer than 15 cases reported in the literature. We report a case of community-acquired Citrobacter koseri spondylodiscitis to highlight the diagnostic and therapeutic management of this unusual pathogen.

Case presentation

A 77-year-old Tunisian woman presented with acute inflammatory dorsal back pain and fever. Laboratory evaluation revealed elevated inflammatory markers. Magnetic resonance imaging (MRI) of the spine demonstrated spondylodiscitis at the T9–T10 level with associated non-compressive epiduritis and paravertebral soft tissue phlegmon. Screening for common infectious foci, including urine cultures and echocardiography, was negative. Blood cultures and a computed tomography (CT)-guided disco-vertebral biopsy both isolated Citrobacter koseri. The patient was successfully treated with a specific antibiotic regimen comprising intravenous ceftazidime and ciprofloxacin for 3 weeks, followed by oral ciprofloxacin and trimethoprim–sulfamethoxazole for a total duration of 3 months. At the 3-month follow-up, the patient had achieved a complete clinical and biological recovery.

Conclusions

This case underscores that Citrobacter koseri can cause spontaneous bacterial spondylodiscitis even in elderly patients from the community with no identifiable urinary or biliary infection. It highlights the importance of definitive microbiological diagnosis via biopsy or blood culture to guide targeted antibiotic therapy. Conservative management with prolonged fluoroquinolone-based regimens can lead to excellent outcomes in patients without neurological deficits.

Background

The genus Citrobacter consists of gram-negative bacilli belonging to the order Enterobacterales, widely distributed in the environment (soil, water, and food) and commonly found commensal inhabitants of the human and animal gastrointestinal tracts [1, 2]. Historically regarded as low-virulence contaminants, Citrobacter species are now recognized as emerging opportunistic pathogens, particularly in hospital settings [3]. Citrobacter freundii and Citrobacter koseri (formerly C. diversus) are the most frequently isolated species, classically associated with urinary tract, intra-abdominal, and wound infections in immunocompromised patients or neonates [4, 5]. Beyond these presentations, C. koseri has been implicated in a broader spectrum of severe invasive infections in adults, including bacteremia, visceral abscesses, pneumorachis, and spinal canal involvement, albeit in exceptionally rare circumstances [2, 5–7].

Infectious spondylodiscitis (ISD) caused by Citrobacter species is an exceptionally rare clinical entity. While vertebral osteomyelitis in adults is predominantly attributable to Staphylococcus aureus or Mycobacterium tuberculosis, gram-negative bacilli, and C. koseri in particular, account for only a small minority of cases [6], mostly reported in patients with significant comorbidities or following invasive spinal procedures [7, 8].

The few documented cases have demonstrated considerable clinical heterogeneity, ranging from emphysematous osteomyelitis [9] and pneumorachis secondary to disseminated infection [7], to infected spinal hematoma of iatrogenic origin [10], underscoring the pathogenic versatility of this organism and the diagnostic challenges it poses in spinal infections.

We report a case of community-acquired bacteraemic spondylodiscitis caused by C. koseri in a 77-year-old immunocompetent woman with no identifiable risk factors or history of invasive procedures. This case is notable for the simultaneous microbiological confirmation from both blood cultures and disco-vertebral biopsy, and for the successful outcome achieved through conservative antibiotic therapy alone, without surgical intervention. It contributes to the sparse existing literature on this topic and aims to improve clinician awareness of C. koseri as a causative agent of vertebral osteomyelitis and guide therapeutic decision-making.

Case description

A 77-year-old Tunisian woman presented with persistent, tolerable dorsal back pain and fever that had started 1 week prior to admission. The patient was a housewife with no relevant exposure to farm or domestic animals and no history of recent travel. She lived independently at home with her family and denied tobacco or alcohol use. There was no recent history of infectious symptoms, trauma, surgeries, spinal injections, or skin lesions in the dorsal region. She had no history of diabetes mellitus, neoplasia, or immunosuppression. She had never been previously hospitalized, and was not receiving any regular medication prior to this episode.

On physical examination, her blood pressure was 120/70 mmHg, her heart rate was 100 bpm, and her temperature was 39 °C. She had mild tenderness upon palpation of the lower thoracic spine. Neurological examination revealed neither sensory nor motor deficits, and deep-tendon reflexes were normal in both legs. There were no cardiac murmurs or stigmata of infective endocarditis. The remainder of the physical examination was unremarkable. A detailed timeline of the patient’s clinical presentation, diagnostic milestones, and therapeutic interventions is summarized in Table 1.

Table 1.

Timeline of the patient’s clinical course

Timepoint	Event
T = − 7 days	Onset of dorsal back pain and fever at home
Day 0 (Admission)	Patient presented to the Emergency Department. Physical exam revealed fever (39 °C) and spinal tenderness. Labs showed elevated CRP (277 mg/L)
Day 1	MRI performed, showing spondylodiscitis at T9–T10
Day 3	CT-guided biopsy performed. Empiric antibiotics withheld pending cultures (patient stable)
Day 5	Blood and biopsy cultures positive for Citrobacter koseri. Targeted therapy started (IV Ceftazidime + Oral Ciprofloxacin)
Day 7	Fever resolved. Pain significantly improved
Day 15	CRP levels normalized (< 6 mg/L)
Day 21	Discharged home. Switched to oral consolidation therapy (Ciprofloxacin + TMP–SMX)
Month 3	Follow-up visit. Complete resolution of symptoms. Antibiotics discontinued

CRP:C-reactive protein; MRI: Magnetic resonance imaging; CT: Computed tomography; IV Intravenous; TMP–SMX: Trimethoprim–sulfamethoxazole

Laboratory findings revealed elevated inflammatory markers: an erythrocyte sedimentation rate (ESR) of 95 mm/h, a C-reactive protein (CRP) level of 277 mg/L, and a white blood cell count (WBC) of 9.73 × 10⁹/L. A thoracolumbar spine X-ray showed degenerative changes. Magnetic resonance imaging (MRI) of the spine revealed findings highly suggestive of spondylodiscitis at the T9–T10 intervertebral space. This included reduced signal intensity on sagittal T1-weighted images of the vertebral body, endplate, and disc, and increased signal intensity on sagittal T2-weighted and fat-suppression images at the T9–T10 level. Axial T2-weighted sequences demonstrated vertebral hypersignal, reflecting bone edema extending to the prevertebral soft tissues, suggesting non-compressive epiduritis (Fig. 1).

Fig. 1 .

Magnetic resonance imaging (MRI) of the thoracic spine showing spondylodiscitis at the T9–T10 level. a Sagittal T1-weighted image reveals reduced signal intensity (arrow) of the vertebral bodies, endplates, and intervertebral disc. b Sagittal T2-weighted fat-suppressed image shows increased signal intensity (arrow) consistent with inflammation. c Axial T2-weighted fat-suppressed sequence demonstrates vertebral hypersignal (arrow) reflecting bone edema, with extension into the prevertebral soft tissues

An infection survey was conducted to identify the source. The chest X-ray and transesophageal echocardiography were normal. Urine culture, brucellosis serology, and tuberculosis assessment were negative. Based on the imaging findings, a CT-guided disco-vertebral biopsy was performed. Histopathological examination revealed acute suppurative spondylodiscitis. Polymerase chain reaction (PCR) for Mycobacterium tuberculosis was negative; however, cultures from the biopsy identified C. koseri. Blood cultures were also positive for C. koseri. A diagnosis of spontaneous C. koseri spondylodiscitis with epidural and paraspinal soft tissue extension was established.

The patient was immobilized with a rigid thoracolumbar corset. Antibiotic susceptibility testing showed the C. koseri strain was resistant to ampicillin, cefalexin, cefoxitin, and nitrofurantoin, but susceptible to gentamicin, imipenem, ciprofloxacin, and ceftazidime. The patient was treated with a dual antibiotic regimen comprising intravenous ceftazidime (2 g three times daily) and oral ciprofloxacin (500 mg three times daily). The fever resolved after 48 h of therapy. This regimen was continued for 3 weeks, followed by oral consolidation therapy combining ciprofloxacin (750 mg twice daily) and trimethoprim–sulfamethoxazole 400/80 mg (2 tablets 3 times a day) for a total treatment duration of three months.

Serial monitoring of inflammatory markers demonstrated a rapid biological response to antibiotic therapy. CRP levels normalized within 2 weeks of treatment initiation, while ESR showed a progressive decline, returning to normal range after two months of therapy.

At the 3-month follow-up visit, the patient had completely recovered and returned to her normal baseline function.

Discussion

Spondylodiscitis caused by Citrobacter species is an exceedingly rare clinical entity, particularly in immunocompetent adults. While Staphylococcus aureus remains the predominant pathogen in vertebral osteomyelitis, gram-negative bacilli account for approximately 13–20% of cases, with Escherichia coli being the most common [6, 11].

C. koseri is an infrequent cause of osteoarticular infection. A comprehensive search of the PubMed/MEDLINE database was conducted in April 2026 using the following search terms: "Citrobacter koseri spondylodiscitis", "Citrobacter koseri vertebral osteomyelitis", "Citrobacter koseri spinal infection", and "Citrobacter diversus spondylodiscitis",

Identified only 12 previously reported adult cases, in the English-language literature. The present case, therefore, adds meaningful data to the limited evidence regarding the clinical features, management, and outcomes of this rare infection. A summary of the clinical characteristics, management strategies, and outcomes of all identified cases, including the present one, is provided in Table 2.

Table 2 .

Comparison of the present case with previously reported cases of Citrobacter koseri spondylodiscitis in adults

Author (year)	Age/Sex	Comorbidities	Spinal level	Probable source	Treatment (Duration)	Outcome
Current Case	77/F	None reported	Thoracic (T9–T10)	Unknown (presumed hematogenous)	Ceftazidime, Ciprofloxacin, TMP–SMX (3 months)	Full recovery
Zhang et al. (2026) [9]	72/Not reported	Malnutrition Anemia Hepatitis B exposure	Lumbar L4–L5	Unknown	Meropenem Levofloxacin > 3months	Favorable at 1-month, then Lost to follow-up
Melani et al. 2025 [20]	39/F	Psoriasis	Lumbar L3–L4	Unknown	Ceftriaxone ciprofloxacin 3 months	Residual lower back pain
Zinabu et al. (2024) [15]	18/M	None	Lumbar (L2–L3)	Unknown (post-trauma)	Cefazolin, Cephalexin (initial failure), then Debridement + Fixation	Clinical improvement
Hamdi et al. 2024 [21]	63/M	Hepatitis C,	Lumbar L3–L4	Unknown	cefotaxime gentamicin, then ofloxacin (10 weeks)	Full recovery
Casas-Martínez et al. (2023) [8]	50/M	DM, HTN	Thoracic (T10–T11)	Unknown	Ertapenem, Vancomycin, Ciprofloxacin (6 weeks)	Full recovery
Wu et al. (2023) [13]	62/M	None	Thoracic (T2–T3)	Urinary Tract Infection	Cefixime 6–8 weeks	Clinical improvement
Ramachandran et al. (2022) [11]	27/M	None	Lumbar (L4–L5)	Unknown	Cefoperazone–sulbactam, Ciprofloxacin (8 weeks)	Healed (bony fusion)
Yamamoto et al. 2013 [7]	63/M	Depression DM sacral decubitus ulcers	Pneumorachis	Unknown (presumed hematogenous)	Vancomycin Meropenem (5 weeks)	Full recovery
Hayati et al. (2012) [6]	75/F	None	Lumbar (L2–L3)	Urinary Tract Infection	Cefuroxime (6 weeks)	Full recovery
Canario et al. (2004) [10]	42, M	Pulmonary Tuberculosis	Thoracic (T9–T10)	implanted prosthetic materials	Levofloxacin (> 4 weeks)	Full recovery
Martinez et al. (2001) [22]	70/M	HTN, COPD Pacemaker	Lumbar (L5–S1)	Hematogenous)	Imipenem Tobramycin Then Ofloxacin 16 weeks	Full recovery
Sotto et al. (1994) [19]	72/M	HTN, Angina	Cervical (C3–C4, C6–C7)	Prostatic surgery	Imipenem, Amikacin (3 months)	Recovered (1-year follow-up)

F: Female; M: Male; DM: Diabetes Mellitus; HTN: Hypertension; TMP–SMX: Trimethoprim–sulfamethoxazole; COPD: Chronic Obstructive Pulmonary Disease; NR: Not reported

In adults, C. koseri infections are typically nosocomial and often associated with significant comorbidities, particularly involving urinary or biliary tracts of immunocompromised hosts [8, 12]. Among the reported cases, a urinary source was documented in only two patients [6, 13]. In our case, extensive investigations, including urine culture, failed to identify a primary focus of infection.

Overall, 8 of 13 patients (61.5%) had at least one underlying comorbidity, such as diabetes mellitus, hypertension, chronic hepatitis, or pulmonary tuberculosis, and approximately one-third (30.8%) were considered immunocompromised. Nevertheless, five cases (38.5%), including ours, were community-acquired in patients with no history of recent hospitalization or invasive procedures.

Regarding spinal localization, the lumbar spine was the most frequently affected region, accounting for more than half of reported cases (7/13, 53.8%), followed by the thoracic spine (4/13, 30.8%), and the cervical spine (1/13, 7.7%). In most cases, the primary source of infection remained unidentified, and hematogenous dissemination was presumed. Transient, asymptomatic gastrointestinal translocation has been proposed as a possible mechanism, particularly in elderly patients with degenerative spinal changes, where vascularized endplates may serve as a nidus for infection [14].

The diagnosis of C. koseri spondylodiscitis presents challenges as the clinical presentation is indolent and indistinguishable from other bacterial etiologies. Our patient presented with classic symptoms of fever and localized back pain, supported by elevated inflammatory markers (CRP and ESR). Importantly, blood cultures were positive for C. koseri, which correlated with the biopsy results. This highlights the utility of blood cultures in gram-negative spondylodiscitis, potentially sparing the need for invasive biopsy if the clinical and radiographic picture is consistent, although tissue sampling remains the gold standard for guiding antimicrobial therapy [15].

The antibiotic regimen was guided by both the susceptibility profile of the isolated strain and the intrinsic resistance characteristics of C. koseri. As a group 3 Enterobacterale, C. koseri harbors an inducible chromosomal AmpC β-lactamase that although expressed at low basal levels, carries a well-documented risk of mutational derepression under prolonged β-lactam monotherapy[16] potentially leading to clinical failure despite initial in vitro susceptibility.

Our isolate was resistant to ampicillin and first-generation cephalosporins, consistent with the known intrinsic resistance pattern of C. koseri while remaining fully susceptible to ceftazidime and ciprofloxacin. It is worth noting that, unlike Citrobacter freundii, which more frequently expresses clinically significant AmpC-mediated resistance to third-generation cephalosporins. C. koseri generally retains susceptibility to these agents [17]. Nevertheless, the emergence of acquired resistance in C. koseri warrants continued microbiological vigilance, and susceptibility testing remains essential to guide therapy in all reported cases.

Following 3 weeks of parenteral therapy, during which the patient achieved rapid defervescence within 48 h and a significant decline in inflammatory markers, treatment was transitioned to an oral consolidation regimen comprising ciprofloxacin and trimethoprim–sulfamethoxazole. This combination was chosen on the basis of confirmed in vitro susceptibility and in accordance with the Infectious Disease specialist's recommendation. Both agents offer excellent oral bioavailability and adequate penetration into osteoarticular tissues, making them well-suited for step-down therapy in vertebral osteomyelitis.

Across reported cases, treatment duration ranged from 6 weeks to more than three months. The most common duration was 6–12 weeks (61.5%), while three patients (23%), including ours, received therapy for at least three months.

There is currently no consensus on the optimal duration of antibiotic therapy for Citrobacter spondylodiscitis. While current guidelines recommend a minimum of 6 weeks for gram-negative vertebral osteomyelitis [18], we opted for a prolonged total course of three months, given the presence of epidural involvement, paraspinal soft tissue extension, and documented bacteraemia.

This approach is consistent with the extended durations reported in the reviewed literature (Table 2), including Sotto et al. [19], and is supported by the complete resolution of symptoms and normalization of inflammatory markers at the 3-month follow-up, confirming the efficacy of this regimen in C. koseri spinal infections when the isolate demonstrates full susceptibility.

Surgical management was required in only one reported case [15], involving post-traumatic infection with initial antibiotic failure. All other patients, including ours, were successfully treated with medical therapy alone, achieving full or near-complete recovery.

Conclusions

Citrobacter koseri is a rare but emerging cause of community-acquired spondylodiscitis. This case illustrates its potential to cause severe spinal infection in elderly patients even in the absence of typical risk factors or urinary sources. It underscores the necessity of obtaining a definitive microbiological diagnosis through blood culture or biopsy—prior to initiating antibiotics in hemodynamically stable patients, given the variable resistance profiles of gram-negative pathogens.

Our findings also support the effectiveness of conservative management with prolonged, susceptibility-guided antibiotic therapy. Fluoroquinolone-based regimens, when the isolate is sensitive and neurological function remains intact, can achieve excellent clinical and functional outcomes without the need for surgical intervention.

A review of the literature identified fewer than 15 reported adult cases of C. koseri spondylodiscitis. The lumbar spine was the most frequently involved site, most patients had at least one underlying comorbidity, and surgery was rarely necessary. Overall, treatment durations ranging from 6 weeks to three months—most commonly based on fluoroquinolones—were associated with favorable outcomes in the vast majority of cases.

Abbreviations

CT

Computed tomography

MRI

Magnetic resonance imaging

CRP

C-reactive protein

ESR

Erythrocyte sedimentation rate

ISD

Infectious spondylodiscitis

WBC

White blood cell count

PCR

Polymerase chain reaction

Author contributions

OJ: conceptualization, data collection, clinical management of the patient, drafting of the manuscript, and final approval of the submitted version. MB: data collection, contribution to the literature review, critical revision of the manuscript, and final approval of the submitted version. RB: data collection, contribution to the diagnostic workup, critical revision of the manuscript, and final approval of the submitted version. MA: contribution to the microbiological analysis and interpretation, critical revision of the manuscript, and final approval of the submitted version. RS: contribution to the radiological interpretation, critical revision of the manuscript, and final approval of the submitted version. SA: contribution to the clinical management and follow-up, critical revision of the manuscript, and final approval of the submitted version. FE: contribution to the literature review and discussion, critical revision of the manuscript, and final approval of the submitted version. MY: supervision, critical revision of the manuscript for important intellectual content, and final approval of the submitted version. All authors read and approved the final version of the manuscript.

Funding

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

Ethical approval was not required for this single case report, in accordance with the institutional/national policy.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interest

The authors declare that they have no competing interest.