Abstract
Background
Campylobacter and Salmonella are leading causes of bacterial gastroenteritis worldwide, yet their comparative impact on acute kidney injury (AKI) remains unexplored.
Methods
We conducted a retrospective cohort study of adults (≥18 years) hospitalized with community-acquired bacterial gastroenteritis at a tertiary center in South Korea (2018–2025). The primary outcome was AKI incidence. Multivariable logistic regression assessed associations between pathogen type (Campylobacter, Salmonella, or Others, including Escherichia coli or Clostridium difficile) and AKI risk (reference: Campylobacter). Secondary outcomes included postdischarge kidney outcomes (acute kidney disease [AKD], incident chronic kidney disease [CKD], and CKD progression).
Results
Among 232 patients (Campylobacter n = 139, Salmonella n = 55, and Others n = 38), AKI incidence differed significantly across groups (26.6%, 70.9%, and 31.6%, respectively; P < .001), with 42% of Salmonella patients developing AKI stages 2 and 3. Salmonella enteritis was independently associated with higher AKI risk than Campylobacter (odds ratio 3.19, 95% confidence interval [CI] 1.23–8.30). Salmonella patients had 46% longer hospital stays (median 8 vs 5 days). In-hospital AKI was associated with postdischarge composite kidney outcome (hazard ratio 3.17, 95% CI 1.02–9.83), particularly during the AKD window (7–90 days postdischarge). Notably, 17% of patients who recovered from AKI before discharge subsequently developed AKD or CKD.
Conclusions
In this retrospective study, hospitalized patients with Salmonella enteritis had a higher risk of AKI than those with Campylobacter; furthermore, those with in-hospital AKI from any cause were more likely to have adverse short-to-intermediate–term postdischarge kidney abnormalities. These findings support pathogen-specific risk stratification and the need for postdischarge monitoring of patients who develop in-hospital AKI.
Keywords: acute kidney injury, Campylobacter, chronic kidney disease, gastroenteritis, Salmonella
In this retrospective cohort of 232 hospitalized adults with bacterial gastroenteritis, Salmonella enteritis was associated with a 3.2-fold higher AKI risk than Campylobacter, and 17% of patients recovering from AKI subsequently developed acute kidney disease or chronic kidney disease, suggesting the potential benefit of continued postdischarge kidney monitoring.
Acute kidney injury (AKI) is a frequent and serious complication of acute infectious enteritis, often resulting from prerenal volume depletion and sepsis-induced tubular injury [1–3]. Among patients hospitalized with acute gastroenteritis, the reported AKI incidence ranges from 13.6% to 28.2%, with substantial associated morbidity and mortality [1, 4–6]. Campylobacter and Salmonella species are the leading causes of foodborne bacterial gastroenteritis worldwide [7]. While both pathogens typically present with febrile diarrhea, their systemic impact, particularly on kidney function, remains insufficiently differentiated.
In a 2-year retrospective study, renal dysfunction was reported to occur in ∼36% of adult patients with Salmonella gastroenteritis [8]. A separate retrospective study reported AKI in ∼3.6% of hospitalized patients with Campylobacter gastroenteritis [9]. However, direct head-to-head comparisons within the same population are lacking. Although definitive pathogen identification continues to rely on microbiological testing, characterizing pathogen-specific differences in AKI risk may improve the early recognition of patients at higher risk for renal complications, thereby informing closer monitoring and timely supportive care. Therefore, we evaluated the risk of AKI and postdischarge kidney outcomes among patients hospitalized with community-acquired enteritis caused by Salmonella, Campylobacter, and other common pathogens (Escherichia coli and Clostridium difficile).
METHODS
Study Design, Population, and Data Collection
This retrospective cohort study was conducted at a single tertiary care center in South Korea. We reviewed medical records of all adult patients (≥18 years) admitted with community-acquired bacterial gastroenteritis between January 2018 and December 2025. Eligible patients were those with laboratory-confirmed infections caused by Campylobacter species, Salmonella species, E. coli, or community-acquired C. difficile. Patients transferred from other hospitals and those with healthcare-associated infections were excluded to ensure accurate assessment of clinical presentation and baseline characteristics. The diagnosis was established based on positive results from stool multiplex polymerase chain reaction (PCR), stool culture, or blood culture [10, 11].
Demographic and clinical data were extracted from electronic medical records, including age, sex, comorbidities (diabetes mellitus, hypertension, cancer, cardiovascular disease, cerebrovascular disease, dementia, and autoimmune disease), vital signs at admission, laboratory parameters (blood urea nitrogen, serum creatinine [Cr], white blood cell [WBC] count, C-reactive protein [CRP], and serum sodium), gastrointestinal symptom severity (initial stool frequency and volume), serum Cr at discharge, and length of hospital stay.
The study was approved by the Institutional Review Board of Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, who waived the requirement for informed consent (SCMC 2026-02-009).
Determination of Baseline Creatinine
Baseline serum Cr was defined as the most recent outpatient value measured between 7 days and 1 year prior to admission [12, 13]. For patients without an available preadmission measurement, 2 alternative approaches were applied: (1) the lowest serum Cr measured during hospitalization (nadir Cr) was used as a surrogate baseline Cr; and (2) baseline Cr was back calculated, assuming a baseline estimated glomerular filtration rate (eGFR) of 75 mL/min/1.73 m2, as recommended by the Kidney Disease: Improving Global Outcomes (KDIGO) guideline [12–14]. Baseline eGFR was calculated using the 2021 Chronic Kidney Disease Epidemiology Collaboration equation [15].
Definition of AKI and Postdischarge Kidney Outcomes
Acute kidney injury was defined according to the KDIGO criteria as an increase in serum Cr ≥0.3 mg/dL within 48 hours or ≥1.5× from baseline and was staged as 1 (1.5−1.91.5× baseline or absolute increase ≥0.3 mg/dL), 2 (2.0–2.9× baseline), or 3 (to ≥3.0× baseline, serum Cr ≥4.0 mg/dL, or the initiation of renal replacement therapy) [16]. Estimated baseline AKI (eAKI), used for sensitivity analysis, was defined using back-calculated baseline Cr, assuming an eGFR of 75 mL/min/1.73 m2 [14].
Postdischarge kidney function was assessed through serial outpatient serum Cr measurements. The composite postdischarge kidney outcome comprised acute kidney disease (AKD), incident chronic kidney disease (CKD), and CKD progression. Acute kidney disease was assessed in patients with follow-up <3 months and defined as an eGFR <60 mL/min/1.73 m2, an eGFR decline ≥50% from baseline, or serum Cr ≥1.5× baseline on at least 1 occasion during follow-up [16]. Incident CKD was defined as an eGFR <60 mL/min/1.73 m2 persisting beyond 3 months in patients without preexisting CKD. Chronic kidney disease progression was defined as an eGFR decline ≥50% from baseline or progression to a lower eGFR category in patients with preexisting CKD.
Statistical Analysis
Continuous variables were expressed as median (interquartile range, IQR) or mean ± standard deviation, as appropriate, and categorical variables as frequencies and percentages. Comparisons among the 3 pathogen groups (Campylobacter, Salmonella, and Others) were performed using the Kruskal–Wallis test for continuous variables and Fisher's exact test for categorical variables.
The independent association between Salmonella enteritis and AKI was assessed by multivariable logistic regression with hierarchical adjustment, using Campylobacter as the reference group. Covariates were selected based on clinical relevance, statistically significant differences across pathogen groups, and considerations of model parsimony given the events-per-variable ratio. Four sequential models were constructed: model 1, unadjusted; model 2, adjusted for demographic, comorbidity, and hemodynamic variables (age, sex, diabetes mellitus, hypertension, cancer, systolic and diastolic blood pressure, heart rate, and body temperature); model 3, additionally adjusted for laboratory parameters (CRP, baseline Cr, hyponatremia [<135 mEq/L], and leukocytosis [WBC >10 000/μL]); and model 4, additionally adjusted for gastrointestinal symptom severity (initial stool frequency and volume). The same hierarchical approach was applied to the eAKI sensitivity analysis.
Length of hospital stay was analyzed using Poisson regression with Campylobacter as the reference, yielding incidence rate ratios (IRRs) with 95% confidence intervals (CIs). Time-to-event analysis from hospital discharge to the first composite kidney event or last outpatient visit was performed using Cox proportional hazards regression, adjusted for age, sex, comorbidities (diabetes mellitus, hypertension, and cancer), baseline Cr, AKI, and pathogen type.
All tests were 2-sided, with P < .05 considered statistically significant. Analyses were performed using STATA version 14.2 (StataCorp LP, College Station, TX, United States).
RESULTS
Baseline Characteristics
A total of 232 patients were enrolled, comprising 139 (59.9%) in the Campylobacter group, 55 (23.7%) in the Salmonella group, and 38 (16.4%) in the Others group. All patients underwent PCR. Blood cultures were obtained prior to antibiotic administration in 225 of 232 patients (97.0%); the remaining 7 patients included 2 in the Campylobacter group and 5 in the Others group. Stool cultures were available in 210 patients (90.5%), with no difference in stool-culture testing across pathogen groups (P = .857). Despite this systematic testing, most Campylobacter (97.0%) cases and all cases in Others group (100%) were diagnosed by PCR alone due to negative cultures. In contrast, Salmonella cases were identified by PCR alone (50.0%), both PCR and culture (30.4%), or culture alone (blood: 10.7%; stool: 8.9%). Blood cultures were positive in 9 of 55 Salmonella patients (16.4%) and 4 of 139 Campylobacter patients (2.9%). Among the 13 E. coli cases, 3 (23.1%) were Shiga toxin–producing E. coli (STEC; 1 E. coli O157 and 2 enterohemorrhagic E. coli); the remainder were enterotoxigenic E. coli (n = 5), enteropathogenic E. coli (n = 2), and enteroaggregative E. coli (n = 3).
A measured preadmission serum Cr (pre-Cr) was available in 142 patients (61.2%); for the remaining 90 (38.8%), the in-hospital nadir Cr was used as the baseline. The median interval between pre-Cr measurement and admission did not differ across pathogen groups (P = .062; Supplementary Table 1). Baseline characteristics by pathogen group are summarized in Table 1. Baseline serum Cr did not differ significantly among the groups, whereas Cr at admission was highest in the Salmonella group (Table 1). In a subgroup analysis restricted to the 142 patients with measured pre-Cr, the median values of pre-Cr, nadir Cr, and back-calculated baseline Cr did not differ across pathogen groups (Supplementary Table 1).
Table 1.
Baseline Characteristics Among 232 Patients Hospitalized With Campylobacter, Salmonella, and Other Bacterial Enteritis
| Pathogen Type | |||||
|---|---|---|---|---|---|
| Total | Campylobacter | Salmonella | Others | P Value | |
| N (%) | 232 | 139 (59.9) | 55 (23.7) | 38 (16.4) | |
| Age, y | 59 ± 20 | 54 ± 21 | 66 ± 13 | 65 ± 18 | <.001 |
| Male, % | 44.4 | 41.7 | 50.9 | 44.7 | .51 |
| Diabetes, % | 22.4 | 15.1 | 41.8 | 21.1 | <.001 |
| Hypertension, % | 37.5 | 36.0 | 45.5 | 31.6 | .334 |
| Cardiovascular disease, % | 11.2 | 12.2 | 7.3 | 13.2 | .563 |
| Cerebrovascular disease, % | 7.8 | 5.8 | 7.3 | 15.8 | .160 |
| Cancer, % | 6.5 | 2.9 | 14.6 | 7.9 | .009 |
| Dementia, % | 2.2 | 2.2 | 1.8 | 2.6 | 1.000 |
| Autoimmune disease, % | 1.7 | 2.2 | 0 | 2.6 | .625 |
| BUN, mg/dL | 15 (11–27) | 14 (10–18) | 29 (20–55) | 15 (11–29) | <.001 |
| Baseline creatinine, mg/dL | 0.7 (0.6–0.9) | 0.7 (0.6–0.8) | 0.8 (0.6–1.0) | 0.7 (0.6–1.0) | .076 |
| Estimated creatinine,a mg/dL | 0.9 (0.8–1.1) | 0.9 (0.8–1.1) | 0.9 (0.7–1.0) | 1.0 (0.7–1.3) | .253 |
| Creatinine at admission, mg/dL | 0.9 (0.7–1.4) | 0.8 (0.7–1.1) | 1.4 (0.9–4.0) | 0.8 (0.6–1.4) | <.001 |
| Leukocytosis, % | 49.1 | 53.2 | 38.2 | 50.0 | .166 |
| C-reactive protein, mg/L | 111 (61–175) | 107 (64–157) | 174 (104–248) | 53 (20–119) | <.001 |
| Serum sodium, mmol/L | 136.0 ± 4.0 | 136.2 ± 3.1 | 134.2 ± 4.7 | 137.6 ± 5.1 | <.001 |
| Hyponatremia, % | 29.7 | 25.9 | 47.3 | 18.4 | .003 |
| SBP, mm Hg | 120 ± 22 | 125 ± 20 | 108 ± 22 | 120 ± 22 | <.001 |
| DBP, mm Hg | 73 ± 15 | 75 ± 13 | 67 ± 18 | 73 ± 15 | <.001 |
| Heart rate, beats/min | 98 (80–109) | 98 (80–110) | 100 (80–110) | 90 (78–100) | .093 |
| Body temperature, °C | 38.0 (37.1–38.8) | 38.3 (37.4–39.1) | 38.0 (37.1–38.9) | 37.0 (36.7–38.1) | <.001 |
| Diarrhea frequency, times/d | 5.0 (3.0–9.5) | 5.0 (2.0–10.0) | 7.0 (4.0–10.0) | 5.0 (3.0–9.0) | .049 |
| Diarrhea amount, g/d | 400 (200–670) | 350 (200–500) | 700 (450–1150) | 350 (200–550) | <.001 |
Values for continuous variables are shown as mean ± standard deviation or median (interquartile range). Values for categorical variables are shown as frequencies (percentages). Leukocytosis was defined as a white blood cell count >9400/μL, and hyponatremia was defined as a serum sodium level <135 mEq/L.
Abbreviations: BUN, blood urea nitrogen; SBP, systolic blood pressure; DBP, diastolic blood pressure.
aEstimated creatinine is the baseline creatinine back calculated, assuming an estimated glomerular filtration rate of 75 mL/min/1.73 m2.
Compared with the other groups, Salmonella patients exhibited significantly higher CRP levels (median 174 mg/L vs 107 and 53 mg/L; P < .001), greater daily stool volume (700 g/day vs 350 and 350 g/day; P < .001), and more frequent diarrhea (7.0 vs 5.0 and 5.0 episodes/day; P = .049). Hyponatremia was predominantly mild (130–134 mmol/L) across all groups, with severe hyponatremia (<125 mmol/L) observed in only 3 patients overall (Supplementary Table 2). In contrast, the Campylobacter group had a higher body temperature at presentation (median 38.3 ℃ vs 38.0 ℃ and 37.0 ℃; P < .001). Bloody diarrhea was uncommon (n = 11; 2 in Campylobacter, 1 in Salmonella, and 8 in Others, predominantly E. coli). Although Salmonella patients were older than the other groups, none of the patients, including those of advanced age, were bedridden or fully dependent for activities of daily living (Supplementary Table 3).
Incidence and Severity of AKI
The incidence of AKI differed significantly across pathogen groups (P < .001), affecting 37 of 139 Campylobacter patients (26.6%), 39 of 55 Salmonella patients (70.9%), and 12 of 38 in the Others group (31.6%). While AKI stage 1 was comparable across groups, the combined incidence of AKI stages 2 and 3 was markedly higher in the Salmonella group (41.8%) than in the Others groups (Figure 1A). Median peak serum Cr in patients with AKI was also highest in the Salmonella group (2.03 vs 1.30 mg/dL in Campylobacter and 1.65 mg/dL in the Others group; P = .0498; Supplementary Table 4). Of the 13 E. coli cases, AKI developed in 4, all caused by non-STEC pathotypes (enterotoxigenic E. coli, n = 2; enteropathogenic E. coli, n = 1; and enteroaggregative E. coli, n = 1). Four patients (1.7%) required transient dialysis (1–2 sessions) during hospitalization (3 in Campylobacter, 1 in Salmonella, and 0 in Others; P = 1.000); all discontinued dialysis after kidney function recovered.
Figure 1.
Incidence and severity of acute kidney injury (AKI) by pathogen group. A, AKI incidence based on the primary analysis. B, Estimated AKI (eAKI) incidence based on the sensitivity analysis using back-calculated baseline creatinine (estimated glomerular filtration rate 75 mL/min/1.73 m2). Proportions of patients with no AKI, AKI stage 1, and AKI stages 2 and 3 are shown for Campylobacter, Salmonella, and other bacterial enteritis (Escherichia coli or Clostridium difficile). Between-group comparisons: *P < .001; **P < .05.
Multivariable logistic regression confirmed that Salmonella enteritis was independently associated with increased AKI risk compared with Campylobacter (adjusted odds ratio [OR] 3.19, 95% CI 1.23–8.30; P = .017) in the fully adjusted model (Figure 2). This association remained robust after additional adjustment for cardiovascular disease, cerebrovascular disease, dementia, and autoimmune disease (Supplementary Table 5).
Figure 2.
Forest plot showing odds ratios and 95% confidence intervals for acute kidney injury risk in Salmonella and Others groups compared with Campylobacter enteritis (reference group). Four sequential models with hierarchical adjustment: model 1 was unadjusted; model 2 was adjusted for age, sex, diabetes mellitus, hypertension, cancer, systolic blood pressure, diastolic blood pressure, heart rate, and body temperature; model 3 was adjusted for model 2 variables plus C-reactive protein, baseline serum creatinine, the presence of hyponatremia, and the presence of leukocytosis; and model 4 was adjusted for model 3 variables plus stool frequency and stool volume.
Most patients received third-generation cephalosporins (73.9%) or piperacillin–tazobactam (13.0%), with adjunctive metronidazole administered in 78.0% of cases (Supplementary Table 6); antibiotic dosing was routinely adjusted for kidney function in patients with AKI. No patient was exposed to nephrotoxic agents, such as nonsteroidal anti-inflammatory drugs, aminoglycosides, or intravenous vancomycin.
Length of Hospital Stay and AKI Recovery Before Discharge
The median length of hospital stay was significantly longer in the Salmonella group (8.0 days, IQR 6.0–11.0) than in the Campylobacter group (5.0 days, IQR 4.0–7.0) and Others group (6.5 days, IQR 5.0–9.0; Table 2). Poisson regression confirmed this difference (unadjusted IRR 1.46, 95% CI 1.31–1.63 for Salmonella and 1.36, 95% CI 1.19–1.56 for Others; both P < .001), with the associations remaining significant after multivariable adjustment (adjusted IRR 1.17, 95% CI 1.03–1.34 for Salmonella and 1.31, 95% CI 1.14–1.50 for Others; Table 2).
Table 2.
Length of Hospital Stay and AKI Recovery Before Discharge by Pathogen Group
| Variable | Campylobacter (n = 139) |
Salmonella
(n = 55) |
Others (n = 38) |
|---|---|---|---|
| Length of hospital stay | |||
| Median (IQR)*, d | 5.0 (4.0–7.0) | 8.0 (6.0–11.0) | 6.5 (5.0–9.0) |
| Range, d | 2–23 | 3–30 | 4–22 |
| Crude IRR (95% CI) | 1.00 (Reference) | 1.46 (1.31–1.63)** | 1.36 (1.19–1.55)** |
| Adjusted IRR (95% CI) | 1.00 (Reference) | 1.17 (1.03–1.34)*** | 1.31 (1.14–1.50)** |
| AKI recovery | |||
| AKI patients, n (%)* | 37 (26.6) | 39 (70.9) | 12 (31.6) |
| AKI recovery, n (%) | 32 (86.5) | 34 (87.2) | 12 (100.0) |
Poisson regression was adjusted for age, sex, diabetes mellitus, hypertension, cancer, blood pressure, heart rate, body temperature, C-reactive protein, baseline serum creatinine, hyponatremia, and leukocytosis. Recovery rate calculated among patients with AKI during hospitalization.
Abbreviations: AKI, acute kidney injury; IRR, incidence rate ratio; IQR, interquartile range.
*P < .001 between groups; **P < .001; ***P = .02.
Despite the longer hospital stays, AKI recovery rate, defined as restoration of eGFR to >90% of baseline value, did not differ across groups (86.5%, 87.2%, and 100% in the Campylobacter, Salmonella, and Others groups, respectively; P = .477; Table 2). No patient required renal replacement therapy beyond brief, transient sessions, indicating that AKI in this cohort, although more severe in Salmonella group, was largely transient and reversible during hospitalization.
Sensitivity Analyses
When baseline Cr was back calculated, assuming an eGFR of 75 mL/min/1.73 m2 (eAKI definition), a similar pattern was observed across groups (P < .001): 20.1% in Campylobacter, 63.6% in Salmonella, and 18.4% in Others (Figure 1B). Multivariable analysis confirmed the independent association between Salmonella enteritis and eAKI risk (adjusted OR 3.70, 95% CI 1.20–11.84; P = .023; Supplementary Table 7).
To address the potential bias from using nadir Cr as a surrogate baseline in 90 patients without a measured pre-Cr, we performed a sensitivity analysis restricted to the 142 patients with available pre-Cr values. The independent association between Salmonella enteritis and AKI risk was preserved (adjusted OR 3.96, 95% CI 1.16–13.49; P = .028), with comparable results when nadir Cr was used as the baseline within this subgroup (adjusted OR 5.43, 95% CI 1.30–22.72; P = .020; Supplementary Table 8). Importantly, baseline characteristics did not differ meaningfully between patients with and without measured pre-Cr (all absolute standardized differences <.2; Supplementary Table 9).
In a sensitivity analysis adding bacteremia to the multivariable model, the association between Salmonella and AKI remained essentially unchanged (adjusted OR 2.97, 95% CI 1.13–7.80; P = .028), whereas bacteremia was not independently associated with AKI (adjusted OR 3.86, 95% CI .59–25.04; P = .157; Supplementary Table 10). Bacteremia was associated with AKI across the cohort (76.9% vs 35.6%; P = .006); among Salmonella patients, AKI occurred in 30 of 46 (65.2%) without bacteremia and in all 9 (100%) with bacteremia (P = .046).
Postdischarge Composite Kidney Outcomes
Of the 232 patients, 194 (83.6%) had at least 1 postdischarge serum Cr measurement, with a median follow-up of 324 days (IRQ 20–864 days). Of these, 66 patients had follow-up of 7–90 days (evaluable for AKD only) and 128 had follow-up ≥90 days (evaluable for incident CKD and CKD progression; Supplementary Table 11). The composite kidney outcome of AKD, CKD, or CKD progression developed in 19 patients (9.8%), and in-hospital AKI was significantly associated with this composite outcome (adjusted hazard ratio [HR] 3.20, 95% CI 1.03–9.93; P = .045; Table 3).
Table 3.
Association of AKI at Admission With Postdischarge Composite Kidney Outcomes: Unadjusted and Adjusted Cox Regression Analysis
| Unadjusted Model OR (95% CI) |
P Value | Adjusted Model OR (95% CI) |
P Value | |
|---|---|---|---|---|
| AKI | 3.11 (1.22–7.92) | .023 | 3.20 (1.03–9.93) | .045 |
| Pathogen type | ||||
| Campylobacter | … | … | 1.0 (reference) | … |
| Salmonella | … | … | 0.61 (.17–2.13) | .438 |
| Others | … | … | 0.38 (.09–1.69) | .204 |
| Age | 1.04 (1.0–1.08) | .051 | ||
| Sex (male) | … | … | 0.39 (.14–1.09) | .072 |
| Diabetes | … | … | 0.79 (.27–2.32) | .673 |
| Hypertension | … | … | 1.59 (.47–5.39) | .459 |
| Cancer | … | … | 1.28 (.25–6.62) | .765 |
| Baseline creatinine | … | … | 1.58 (.88–2.84) | .124 |
Composite kidney outcomes were composed of acute kidney disease, incident chronic kidney disease, and chronic kidney disease progression.
Abbreviations: AKI, acute kidney injury; CI, confidence interval; OR, odds ratio.
When analyzed by outcome type and follow-up window, the strongest signal was observed in the AKD window: AKD developed in 6 of 28 patients with AKI (21.4%) compared with 1 of 38 patients without AKI (2.6%; P = .036; Supplementary Figure 1). The longer-term de novo CKD and combined CKD outcomes did not reach statistical significance but showed point estimates directionally consistent with this association (HR >1.0; Supplementary Table 10). However, post-90-day follow-up data were not available for 7 patients who developed AKD within 7–90 days of discharge, precluding determination of whether their AKD progressed to CKD or resolved. Notably, among the 65 patients who recovered from AKI before discharge with available follow-up data, 11 (16.9%) subsequently developed AKD or CKD, highlighting the importance of continued kidney function monitoring even after apparent recovery.
DISCUSSION
This retrospective cohort study demonstrates that Salmonella enteritis is associated with a significantly higher incidence and severity of AKI compared with Campylobacter enteritis and other bacterial gastroenteritis. Among 232 hospitalized patients with community-acquired bacterial gastroenteritis, Salmonella infection was associated with a 3.2-fold higher risk of AKI even after adjusting for demographics, hemodynamic parameters, inflammatory markers, and gastrointestinal symptom severity. Notably, 42% of Salmonella patients developed AKI stage 2 or 3, compared with significantly lower rates in other pathogens. Given the retrospective design and greater baseline illness severity in the Salmonella group, this association likely reflects a marker of, and contributor to, overall disease severity rather than a purely pathogen-specific effect. These findings remained consistent in sensitivity analyses using different baseline Cr definitions. Furthermore, in-hospital AKI was independently associated with adverse postdischarge kidney outcomes, primarily in the short-to-intermediate term (AKD). Although longer-term CKD associations were not statistically significant, they were directionally consistent, suggesting that patients with in-hospital AKI require continued postdischarge monitoring, even if they appear to have recovered.
The higher AKI incidence and severity observed in Salmonella infection should be interpreted in the context of the clinical course in our cohort: most patients recovered kidney function rapidly with supportive care—often within 48 hours of admission—(recovery rates 86.5%–100%), and only one Salmonella patient required transient dialysis. This pattern is most consistent with functional, hemodynamically mediated AKI rather than persistent structural injury [8]. The persistent association after adjustment for stool frequency and volume suggests that pathogen-related factors—including the more pronounced systemic inflammatory response, greater third-spacing from mucosal edema, and a higher rate of transient bacteremia in Salmonella (16.4% vs 2.9% in Campylobacter)—may amplify the magnitude of this functional injury [8, 17]. In addition, the higher AKI risk in Salmonella enteritis was not explained by bacteremia alone: even nonbacteremic Salmonella patients had a substantially higher AKI rate (65.2%) than the overall Campylobacter group (26.6%), and adjustment for bacteremia did not materially alter the association. This suggests that bacteremia acts as a partial mediator and marker of disease severity rather than the sole driver of the pathogen-specific association, though the small number of bacteremic patients (n = 13) limits precision. Salmonella type III secretion system effectors, such as SipA, SopB, and SopE2, enhance mucosal inflammation and vascular permeability, which may indirectly impair renal perfusion through cytokine-mediated mechanisms, although severe parenchymal complications, such as glomerulonephritis or interstitial nephritis described in case reports, were not observed in our cohort [17, 18]. Consistent with this greater systemic burden, hyponatremia was nearly twice as common in Salmonella (47.3% vs 25.9% in Campylobacter), likely reflecting the combined effects of greater intestinal sodium losses and nonosmotic vasopressin release secondary to effective volume depletion and systemic inflammation.
However, the high rate of apparent recovery should not be interpreted as a complete return to baseline kidney health. Restoration of GFR does not necessarily reflect complete tubular regeneration: some injured tubular cells die and are not regenerated, with eGFR normalization occurring through compensatory hypertrophy of the remaining cells—a process that constitutes a risk factor for subsequent CKD [19]. Consistent with this, in-hospital AKI in our cohort was associated with a significant signal for postdischarge AKD, supporting the recommendation that patients who experience AKI during acute gastroenteritis, particularly those with severe (stages 2 and 3) AKI, warrant continued kidney function monitoring after discharge, even when apparent recovery has occurred.
Our findings align with and extend previous observations regarding pathogen-specific differences in AKI risk during bacterial gastroenteritis. Janssen van Doorn et al reported that 36% of adult patients with Salmonella gastroenteritis developed acute renal failure, compared with only 5% with other gastrointestinal infections [8]. A recent tertiary care study from Saudi Arabia identified C. difficile gastroenteritis as the bacterial enteric infection most strongly associated with AKI [1]. However, that cohort included a high proportion of patients with cancer (47%) and predominantly hospital-acquired infections, fundamentally differing from our community-acquired adult population.
Our study extends prior work in several important ways. First, to our knowledge, this is the first direct head-to-head comparison of Campylobacter and Salmonella—the most common community-acquired bacterial gastroenteritis pathogens worldwide—within the same population. This is clinically meaningful because both pathogens present with similar initial symptoms (febrile diarrhea), yet our data demonstrate strikingly different renal risks that may warrant distinct management approaches. Second, we validated our findings across multiple sensitivity analyses using different baseline Cr definitions (measured preadmission values, in-hospital nadir values, and back-calculated values), with concordant results supporting robustness against baseline Cr misclassification—a common concern in retrospective AKI studies. Third, by systematically excluding transferred and hospital-acquired cases, we focused on a homogeneous community-acquired population, reducing confounding from nosocomial factors. Finally, we provide novel data on postdischarge kidney outcomes, with 17% of patients who recovered from AKI before discharge subsequently developing AKD or CKD.
Several limitations should be acknowledged. First, as a retrospective single-center study, our findings may not be generalizable to other settings with different pathogen distributions or clinical practice. In particular, the preponderance of Campylobacter likely reflects single-center referral bias, case-mix patterns, and the high sensitivity of PCR for Campylobacter rather than true population incidence. This sample imbalance reduced the statistical power for comparisons with the smaller Salmonella and Others groups. Second, baseline Cr values prior to admission were available in only 142 of 232 patients (61.2%); for the remaining 90 patients, the in-hospital nadir Cr was used as the baseline. We addressed this through sensitivity analyses using back-calculated baseline Cr and analyses restricted to patients with measured preadmission Cr, all of which yielded directionally consistent results. Third, certain potentially relevant variables (eg, creatine kinase levels, detailed fluid resuscitation parameters, or biomarkers of tubular injury) and detailed preadmission functional status were not systematically captured, which may limit mechanistic interpretation. Fourth, the modest sample size—particularly in the Salmonella and Others groups—limits the precision of our effect estimates and subgroup analyses. Fifth, postdischarge follow-up duration varied across patients. Specifically, 7 patients who developed AKD within 7–90 days lacked follow-up data beyond 90 days, and the longer-term CKD analyses were underpowered, as reflected in the wide CIs; consequently, the direction of any resulting bias cannot be determined. Finally, our study included only hospitalized patients, and the AKI risk among outpatients with milder gastroenteritis remains unknown. Given the retrospective design and modest sample size, prospective multicenter studies with standardized postdischarge follow-up are needed to confirm the pathogen-specific risk patterns observed here and to better characterize the long-term kidney consequences of in-hospital AKI.
In conclusion, in this retrospective cohort study of hospitalized adults with community-acquired bacterial gastroenteritis, Salmonella enteritis was associated with a 3.2-fold increased risk of AKI compared with Campylobacter enteritis, likely reflecting both pathogen-related factors and greater overall disease severity, with nearly half of Salmonella patients developing AKI stages 2 and 3. In-hospital AKI, regardless of the causative pathogen, was further associated with adverse short-to-intermediate–term postdischarge kidney outcomes. These findings support pathogen-specific risk stratification—specifically, closer renal monitoring and prompt fluid resuscitation in patients with Salmonella infection—and highlight the need for continued postdischarge kidney function monitoring for those who develop in-hospital AKI.
Supplementary Material
Notes
Data availability. All data generated or analyzed during this study are included in this published article.
Contributor Information
Hyeon Woo, Division of Nephrology, Department of Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
Woogyeom Kim, Division of Nephrology, Department of Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
Yu Mi Wi, Division of Infectious Diseases, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
Dae Hyeon Cho, Division of Gastroenterology and Hepatology, Department of Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
Yu-Ji Lee, Division of Nephrology, Department of Medicine, Samsung Changwon Hospital, Sungkyunkwan University School of Medicine, Changwon, Republic of Korea.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. Bogari MH, Munshi A, Almuntashiri S, et al. Acute gastroenteritis-related acute kidney injury in a tertiary care center. Ann Saudi Med 2023; 43:82–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Haridas N, Thirumavalavan S, Fernando ME, et al. Post-diarrheal acute kidney injury during an epidemic in monsoon—a retrospective study from a tertiary care hospital. Indian J Nephrol 2024; 34:338–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Bhakthavatchalam S, Srinivasan D, Prithviraj R. The requirement of hemodialysis in patients with acute gastroenteritis-induced acute kidney injury. J Family Med Prim Care 2021; 10:2423–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Marzuillo P, Baldascino M, Guarino S, Perrotta S, Miraglia Del Giudice E, Nunziata F. Acute kidney injury in children hospitalized for acute gastroenteritis: prevalence and risk factors. Pediatr Nephrol 2021; 36:1627–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Sahay M. The ongoing saga of acute kidney injury associated with gastroenteritis in developing world. Indian J Nephrol 2024; 34:279–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bradshaw C, Zheng Y, Silver SA, Chertow GM, Long J, Anand S. Acute kidney injury due to diarrheal illness requiring hospitalization: data from the national inpatient sample. J Gen Intern Med 2018; 33:1520–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kaakoush NO, Castano-Rodriguez N, Mitchell HM, Man SM. Global epidemiology of Campylobacter infection. Clin Microbiol Rev 2015; 28:687–720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Janssen van Doorn K, Pierard D, Spapen H. Acute renal dysfunction in Salmonella gastroenteritis. J Clin Gastroenterol 2006; 40:910–2. [DOI] [PubMed] [Google Scholar]
- 9. Moffatt CRM, Kennedy KJ, Selvey L, Kirk MD. Campylobacter-associated hospitalisations in an Australian provincial setting. BMC Infect Dis 2021; 21:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Shane AL, Mody RK, Crump JA, et al. 2017 Infectious Diseases Society of America clinical practice guidelines for the diagnosis and management of infectious diarrhea. Clin Infect Dis 2017; 65:e45–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Humphries RM, Linscott AJ. Practical guidance for clinical microbiology laboratories: diagnosis of bacterial gastroenteritis. Clin Microbiol Rev 2015; 28:3–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Khwaja A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clin Pract 2012; 120:c179–84. [DOI] [PubMed] [Google Scholar]
- 13. Siew ED, Ikizler TA, Matheny ME, et al. Estimating baseline kidney function in hospitalized patients with impaired kidney function. Clin J Am Soc Nephrol 2012; 7:712–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Pickering JW, Endre ZH. Back-calculating baseline creatinine with MDRD misclassifies acute kidney injury in the intensive care unit. Clin J Am Soc Nephrol 2010; 5:1165–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Inker LA, Eneanya ND, Coresh J, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med 2021; 385:1737–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Palevsky PM, Liu KD, Brophy PD, et al. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for acute kidney injury. Am J Kidney Dis 2013; 61:649–72. [DOI] [PubMed] [Google Scholar]
- 17. Worley MJ. Salmonella type III secretion system effectors. Int J Mol Sci 2025; 26:2611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Caers J, Peeters P, Vanden Houte K, Sennesael J, Van der Niepen P, Verbeelen D. Acute interstitial nephritis associated with salmonellosis. Eur J Intern Med 2006; 17:217–9. [DOI] [PubMed] [Google Scholar]
- 19. Goldstein SL, Chawla L, Ronco C, Kellum JA. Renal recovery. Crit Care 2014; 18:301. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.