Penicillin-nonsusceptible Streptococcus suis infections in humans: clinical characteristics and antimicrobial synergy testing Authors: Wilawan Thipmontree, Wiyada Chumpol, Kamonwan Lunha, Rattagan KajeeKul, Anusak Kerdsin, Suganya Yongkiettrakul Published: 2026 ABSTRACT BackgroundAntimicrobial resistance in Streptococcus suis represents an increasing global health concern, particularly in Thailand, where human and swine infections remain common and severe.MethodsA descriptive study was conducted in northeastern Thailand from January 2020 to December 2024 to examine the association between clinical characteristics and penicillin susceptibility (susceptible, intermediate, resistant) among S. suis isolates. In vitro antibiotic interactions were assessed by checkerboard microdilution in 23 drug-resistant isolates.ResultsAmong 373 culture-confirmed S. suis cases, 161 isolates (43.2%) were analyzed for serotype and clonal complex (CC); serotype 2 predominated (95.0%), and CC1 was most common (30.4%). Most patients were male (70.2%), with a mean age of 56.9 ± 13.8 years and median illness duration of 3 days (IQR, 1–10). Bacteremia was the predominant clinical manifestation, followed by infective endocarditis (IE) and meningitis. Penicillin resistance increased from 7.1% in 2020 to 57.8% in 2024, whereas ceftriaxone resistance remained infrequent. Of 373 isolates, 62 were penicillin-susceptible, 225 intermediate, and 86 resistant. Meningitis was more frequently observed in penicillin-susceptible infections, whereas IE predominated in penicillin-nonsusceptible infections (P < 0.001). The overall case fatality rate was 28.4% (95% CI, 23.8–33.3) and did not differ by penicillin susceptibility group (P = 0.710). In vitro synergism was most frequently observed for ampicillin–ceftriaxone (39.1%) and penicillin–ceftriaxone (21.7%), whereas gentamicin combinations were mostly indifferent.ConclusionPenicillin-nonsusceptible S. suis was frequently identified among infections in northeastern Thailand. In vitro synergistic activity was observed for the ampicillin–ceftriaxone combination in a subset of isolates, warranting further investigation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12879-026-13332-4. Introduction Streptococcus suis is a zoonotic, gram-positive bacterium classified into 29 serotypes based on its capsular polysaccharide antigens, with serotype 2 being the most prevalent [1, 2]. S. suis commonly inhabits the upper respiratory tract of pigs and is capable of infecting both swine and humans worldwide [1, 3]. Human infections typically occur through the consumption of raw pork products or occupational contact with S. suis–infected pigs [4]. The extensive use of antibiotics in the pig industry to control S. suis infections is considered a potential contributor to the global rise in antimicrobial resistance (AMR). Patterns of antibiotic use in pig production vary widely across farms, countries, and regions, which may partly account for differences in AMR rates across antibiotic classes in Europe, the Americas, and Asia [5]. A previous study by Yongkiettrakul et al. [6], conducted between 2006 and 2015 in northern, central, and southern Thailand, examined S. suis isolates from diseased pigs, healthy pigs, and humans. The study reported widespread multidrug resistance, particularly to lincosamides, macrolides, clindamycin, and tetracycline. Notably, penicillin-nonsusceptible isolates were identified in 68.8% of healthy pigs, 28.3% of diseased pigs, and none of the human isolates. From 2018 to 2020, Lunha et al. analyzed samples from sick pigs across Thailand and reported high levels of reduced susceptibility to ampicillin, penicillin, cefotaxime, ceftriaxone, ceftiofur, chloramphenicol, gentamicin, florfenicol, and tiamulin. In contrast, most isolates remained susceptible to amoxicillin/clavulanic acid, ertapenem, meropenem, and vancomycin [7]. In 2021, human S. suis isolates in Thailand showed intermediate resistance to penicillin (8.2%) while remaining fully susceptible to ceftriaxone [8]. However, recent studies have reported a rapid rise in penicillin-nonsusceptible isolates [9, 10]. Decreased susceptibility to β-lactams is mainly associated with amino acid substitutions in penicillin-binding proteins (PBPs), which reduce drug-binding affinity. The cumulative acquisition of PBP variants contributes to incremental increases in minimum inhibitory concentrations (MICs) [9, 11], and emerging ceftriaxone nonsusceptibility may further increase morbidity and mortality [10]. As a zoonotic pathogen, the increasing AMR of S. suis in both pigs and humans represents a significant public health concern in Thailand. Combination therapy may offer a promising approach for managing resistant bacterial infections, with the potential to enhance bacterial killing and limit the emergence of resistance [12]. Studies of S. suis isolates from swine have demonstrated synergistic effects with combinations such as ampicillin plus apramycin and spectinomycin plus tiamulin [13]. However, research on antimicrobial synergy in human isolates remains limited. This study examined the clinical characteristics associated with S. suis infections caused by penicillin-susceptible, -intermediate, and -resistant strains and evaluated in vitro antimicrobial synergy using a checkerboard microdilution assay. Materials and methods Study setting and design This descriptive study included patients with S. suis infection admitted to MNRH, a 1,500-bed referral center in northeastern Thailand, between January 1, 2020, and December 31, 2024. Infection was defined by the isolation of S. suis from blood, cerebrospinal fluid, synovial fluid, pus, or tissue; multiple positive cultures from the same patient were considered a single case. Species identification was performed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS; Bruker Biotyper, Bruker Daltonics, Germany) according to the manufacturer’s instructions, with a score ≥ 2.0 indicating species-level identification. Antimicrobial susceptibility testing was performed on the first isolate from each patient. Definitions S. suis bacteremia was defined as the isolation of S. suis from blood cultures, and complicated bacteremia was defined as bacteremia with involvement of at least one additional organ. Meningitis was defined as cerebrospinal fluid (CSF) inflammation with the isolation of S. suis from CSF. S. suis infective endocarditis was diagnosed according to the 2015 European Society of Cardiology criteria [14]. Septic arthritis and spondylodiscitis were defined by the isolation of S. suis from joint fluid and spinal specimens, respectively. Penicillin susceptibility was classified according to the Clinical and Laboratory Standards Institute (CLSI) guidelines [15] as penicillin-susceptible (PEN-S), penicillin-intermediate (PEN-I), or penicillin-resistant (PEN-R). Penicillin-nonsusceptible (PEN-NS) isolates were defined as those with intermediate or resistant susceptibility (PEN-I or PEN-R). Combination antimicrobial therapy was defined as the concurrent administration of two antimicrobial agents for ≥ 14 consecutive days in patients with a hospital stay exceeding 14 days. Shock was defined as a systolic blood pressure < 90 mmHg. Respiratory failure was defined as the requirement for ventilatory support. Acute kidney injury was defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria as an increase in serum creatinine to ≥ 1.5 times the baseline value or a creatinine level > 2 mg/dL on admission [16]. Disseminated intravascular coagulopathy (DIC) was defined as the presence of coagulopathy accompanied by thrombocytopenia. Multiorgan dysfunction (MOD) was defined as dysfunction of two or more organ systems. Antimicrobial susceptibility testing Antimicrobial susceptibility testing of S. suis isolates was performed in the microbiology laboratory at MNRH. S. suis colonies grown on sheep blood agar after 18 h of incubation at 37 °C in 5% CO₂ were suspended in Mueller–Hinton (MH) broth, and turbidity was adjusted to the 0.5 McFarland standard. The suspension was then inoculated onto MH agar supplemented with sheep blood using a three-dimensional streaking method, after which E-test strips were applied to the agar surface. Plates were incubated for 18 h at 37 °C in 5% CO₂. Antimicrobial susceptibility of S. suis isolates was interpreted according to the CLSI M100 guidelines (34th edition, 2024) [15], using criteria established for the viridans group streptococci [7–9, 15, 17, 18]. For isolates included in antimicrobial synergy testing (n = 23), minimum inhibitory concentrations (MICs) for penicillin, ceftriaxone, ampicillin, and gentamicin were determined by broth microdilution. These MICs were used to calculate the fractional inhibitory concentration index (FICI) using the checkerboard method. For the remaining isolates, penicillin MICs were determined using the E-test (bioMérieux SA, Marcy-l’Étoile, France), with interpretive breakpoints defined as ≤ 0.12 µg/mL (susceptible), 0.25–2 µg/mL (intermediate), and ≥ 4 µg/mL (resistant). Ceftriaxone susceptibility was assessed by disk diffusion in 2020 and by E-test (bioMérieux) in subsequent years, using breakpoints of ≤ 1 µg/mL (susceptible), 2 µg/mL (intermediate), and ≥ 4 µg/mL (resistant). Vancomycin susceptibility was evaluated by disk diffusion (Thermo Scientific™ Oxoid™, Basingstoke, UK). Serotyping and molecular epidemiology Serotyping and clonal complex (CC) analysis were performed on isolates collected since May 2021 at the Faculty of Public Health, Kasetsart University. Isolates that failed to grow in culture were not subjected to PCR; consequently, 161 of the 373 isolates were available for analysis. Multiplex PCR was performed to identify S. suis and its serotypes independently in the four sets, following a previously described method elsewhere [19]. In addition, mismatch amplification mutation assay (MAMA)-PCR, which differentiates between serotypes 1/2 and 2 as well as between serotypes 1 and 14, was applied [20]. Clonal complexes were analyzed by PCR as previously described [21], allowing identification of CC1, CC25, CC28, CC104, and CC233. Isolates that were PCR-negative for these CCs were classified as unlisted. Twenty-one isolates were previously characterized by multilocus sequence typing (MLST) in an earlier study [22]; the remaining isolates in the present study were not subjected to MLST. Checkerboard microdilution assay Synergistic effects of antimicrobial combinations were evaluated using the checkerboard microdilution method in 26 randomly selected S. suis isolates with varying penicillin and ceftriaxone susceptibilities that had undergone serotyping and clonal complex analysis. Three isolates failed to grow under the test conditions; therefore, synergy testing was completed for the remaining 23 isolates. No biological or technical replicates were performed. To ensure assay reliability, each run included appropriate internal quality control strains and reference standards. All procedures followed standardized protocols at the National Center for Genetic Engineering and Biotechnology. The checkerboard microdilution assay was modified from previously described methods [13] to evaluate synergistic interactions among all possible combinations of penicillin G (PEN), ampicillin (AMP), ceftriaxone (CRO), and gentamicin (GEN). In the checkerboard assay, two antimicrobial agents were combined in a two-dimensional dilution matrix. Twofold serial dilutions of the first agent were prepared along the vertical axis of a 96-well microtiter plate, while twofold serial dilutions of the second agent were prepared along the horizontal axis. Each well contained concentrations of both agents ranging from 1/8× MIC to 4× MIC. Briefly, 50 µL of each antibiotic dilution was dispensed into the wells, resulting in a final volume of 100 µL per well in cation-adjusted Mueller–Hinton broth (CAMHB). Each well was inoculated with 50 µL of bacterial suspension at approximately 10⁶ CFU/ml. Positive growth controls consisted of wells without antibiotics, whereas negative controls included wells containing only CAMHB and antibiotic dilutions without inoculum. Synergistic activity was determined by the absence of visible growth after 24 h of incubation at 37 °C in 5% CO₂. The fractional inhibitory concentration index (FICI) was calculated to assess the interaction between antibiotic pairs as follows:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{aligned}\mathrm{F}\mathrm{I}\mathrm{C}\mathrm{I}=\: &\frac{\mathrm{M}\mathrm{I}\mathrm{C}\:\mathrm{o}\mathrm{f}\:\mathrm{A}\:\mathrm{i}\mathrm{n}\:\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{b}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\:\mathrm{M}\mathrm{I}\mathrm{C}\:\mathrm{o}\mathrm{f}\:\mathrm{A}\:\mathrm{a}\mathrm{l}\mathrm{o}\mathrm{n}\mathrm{e}}\cr &+\frac{\mathrm{M}\mathrm{I}\mathrm{C}\:\mathrm{o}\mathrm{f}\:\mathrm{B}\:\mathrm{i}\mathrm{n}\:\mathrm{c}\mathrm{o}\mathrm{m}\mathrm{b}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}{\mathrm{M}\mathrm{I}\mathrm{C}\:\mathrm{o}\mathrm{f}\:\mathrm{B}\:\mathrm{a}\mathrm{l}\mathrm{o}\mathrm{n}\mathrm{e}}\end{aligned}$$\end{document} For each isolate, the lowest FICI value among all wells showing no growth along the growth/no-growth interface was selected as the representative FICI, following the approach described in the Clinical Microbiology Procedures Handbook [23]. Synergism was defined as FICI ≤ 0.5, indifference as 0.5 < FICI ≤ 4, and antagonism as FICI > 4 [12, 24]. Statistical analysis Descriptive statistics were used to summarize baseline characteristics, including means (± standard deviation) and percentages. Comparisons between groups for categorical variables were performed using Fisher’s exact test. A two-tailed P value < 0.05 was considered statistically significant. Results Between 2020 and 2024, 373 culture-confirmed cases of S. suis were identified. Most patients were male (262/373, 70.2%), with a mean age of 56.9 ± 13.8 years. The median duration of illness was 3 days (interquartile range: 1–10 days). Bacteremia was the most frequent clinical manifestation, followed by infective endocarditis (IE) and meningitis. Multiple sites of infection were observed in 44 patients (11.8%), most commonly involving bacteremia, IE, and meningitis (Fig. 1). Rare clinical manifestations included retropharyngeal abscess and gangrenous cholecystitis (one case each), confirmed by pus culture. Fig. 1Clinical manifestations of patients with Streptococcus suis infection (n = 373), 2020–2024. Bacteremia was the most common presentation, followed by infective endocarditis and meningitis Clinical manifestations of patients with Streptococcus suis infection (n = 373), 2020–2024. Bacteremia was the most common presentation, followed by infective endocarditis and meningitis Serotyping and clonal complex (CC) analyses were performed on 161 of 373 isolates (43.2%). Serotype 2 predominated (153/161, 95.0%), followed by serotype 14 (7/161, 4.4%) and serotype 5 (1/161, 0.6%). Among serotype 14 infections, clinical manifestations included meningitis (5/7, 71.4%), IE (1/7, 14.3%), and primary bacteremia (1/7, 14.3%); the single serotype 5 isolate was associated with IE. CC1 was the most prevalent clonal complex (49/161, 30.4%), and the distribution of CCs is shown in Table 1. Penicillin MICs among typed isolates varied across serotypes and CCs. Serotype 2 isolates—particularly CC1, CC1688, and unlisted CCs—exhibiting a wide range of susceptibilities (0.03–8 µg/mL) according to CLSI criteria (Table 1). Table 1Distribution of serotypes, clonal complexes, sequence types, and penicillin MICs among 161 S. suis isolatesSerotypeClonal complexSequenceType aNo. of isolatesPenicillin MIC rangeb (µg/ml)21NA470.03–825NA30.03104NA40.06–2233NA90.03–216881688180.25–41688275622–4UnlistedcNA700.03–8141NA20.25–0.5233NA10.51688168812UnlistedcNA315UnlistedcNA14Abbreviations: MIC, minimal inhibitory concentration; NA, not applicablea Sequence type information for 21 isolates was obtained from a previous study [12]b MIC interpretation: Penicillin MICs were classified as susceptible (≤ 0.12 µg/mL), intermediate (0.25–2 µg/mL), or resistant (≥ 4 µg/mL) according to CLSI M100 (34th ed, 2024)c Unlisted CCs: isolates not belonging to CC1, CC25, CC104, CC233, and CC28; multilocus sequence typing (MLST) was not performed Table 1Distribution of serotypes, clonal complexes, sequence types, and penicillin MICs among 161 S. suis isolatesSerotypeClonal complexSequenceType aNo. of isolatesPenicillin MIC rangeb (µg/ml)21NA470.03–825NA30.03104NA40.06–2233NA90.03–216881688180.25–41688275622–4UnlistedcNA700.03–8141NA20.25–0.5233NA10.51688168812UnlistedcNA315UnlistedcNA14Abbreviations: MIC, minimal inhibitory concentration; NA, not applicablea Sequence type information for 21 isolates was obtained from a previous study [12]b MIC interpretation: Penicillin MICs were classified as susceptible (≤ 0.12 µg/mL), intermediate (0.25–2 µg/mL), or resistant (≥ 4 µg/mL) according to CLSI M100 (34th ed, 2024)c Unlisted CCs: isolates not belonging to CC1, CC25, CC104, CC233, and CC28; multilocus sequence typing (MLST) was not performed Distribution of serotypes, clonal complexes, sequence types, and penicillin MICs among 161 S. suis isolates Abbreviations: MIC, minimal inhibitory concentration; NA, not applicable a Sequence type information for 21 isolates was obtained from a previous study [12] b MIC interpretation: Penicillin MICs were classified as susceptible (≤ 0.12 µg/mL), intermediate (0.25–2 µg/mL), or resistant (≥ 4 µg/mL) according to CLSI M100 (34th ed, 2024) c Unlisted CCs: isolates not belonging to CC1, CC25, CC104, CC233, and CC28; multilocus sequence typing (MLST) was not performed Susceptibility patterns of S. suis to penicillin (PEN) and ceftriaxone (CRO) among the 373 isolates are shown in Fig. 2A and B. The proportion of penicillin-resistant isolates increased over time, from 7.1% in 2020 to 57.8% in 2024, whereas resistance to ceftriaxone remained uncommon throughout the study period. The MIC distribution of penicillin is presented in Additional file 1: Table S1. All S. suis isolates were susceptible to vancomycin by disk diffusion testing. Fig. 2Antimicrobial susceptibility patterns of Streptococcus suis isolates (n = 373) from 2020 to 2024. (A) Proportion of penicillin-resistant isolates, showing an increasing trend over time. (B) Proportion of ceftriaxone-resistant isolates, which remained low throughout the study period Antimicrobial susceptibility patterns of Streptococcus suis isolates (n = 373) from 2020 to 2024. (A) Proportion of penicillin-resistant isolates, showing an increasing trend over time. (B) Proportion of ceftriaxone-resistant isolates, which remained low throughout the study period Patients were classified according to penicillin susceptibility into penicillin-susceptible (PEN-S, n = 62), penicillin-intermediate (PEN-I, n = 225), and penicillin-resistant (PEN-R, n = 86) groups; baseline characteristics are summarized in Table 2. Meningitis was more frequently associated with PEN-S strains than with PEN-I and PEN-R strains (P < 0.001). Transthoracic echocardiography (TTE) was not performed in 72 patients because of the absence of bacteremia (6 cases), death within 7 days of admission (37 cases), or clinical improvement followed by referral to district hospitals (29 cases). Among the 301 patients with S. suis bacteremia who underwent TTE, IE was identified in 51.8% (156/301) and was significantly more prevalent in infections caused by PEN-I and PEN-R strains than in those caused by PEN-S strains (P < 0.001). Table 2Baseline characteristics of 373 S. suis infection casesCharacteristicsTotal(n = 373)PEN-S(n = 62)PEN-I(n = 225)PEN-R(n = 86)P valueDemographics data, n (%) Age ≥ 60 years158 (42.4)28 (45.2)94 (41.8)36 (41.9)0.892 Male262 (70.2)45 (72.6)156(69.3)61 (70.9)0.899 DOI ≥ 7 days129 (34.6)2 (3.2)81 (36.0)46 (53.5)< 0.001 Consume/contact raw pork products60 (16.1)10 (16.1)31 (13.8)19 (22.1)0.202Comorbidities, n (%) Alcoholism62 (16.6)10 (16.1)37 (16.4)15 (17.4)0.980 Diabetes mellitus42 (11.3)10 (16.3)21 (9.3)11 (12.8)0.026Clinical manifestations, n (%) Bacteremia alone96 (25.7)19 (30.6)64 (28.4)13 (15.1)0.030 Complicated bacteremiaa271 (72.4)42 (67.8)157 (69.8)72 (83.7) • Meningitis102 (27.4)32 (51.6)60 (26.7)10 (11.6)< 0.001 • IEb129/3012/44 (4.5)74/178 (41.6)53/79 (67.1)< 0.001 • IEb + Meningitis27/3013/44 (6.8)17/178 (9.6)7/79 (8.9)0.872 • Spondylodiscitis12 (10.9)4 (17.4)6 (8.4)2 (12.5)0.315 No Bacteremia6 (1.6)1 (1.6)4 (1.8)1 (1.2)0.928Complications, n (%) Respiratory failure108 (29.0)17 (27.4)62 (27.7)29 (33.7)0.551 Shock80 (21.4)9 (14.5)52 (23.1)19 (22.1)0.347 AKI (Cr ≥ 2 mg/dL)57 (15.3)8 (12.9)35 (15.6)14 (16.3)0.861 DIC44 (11.8)5 (8.1)30 (13.3)9 (10.5)0.527 Multiorgan dysfunction83 (22.3)9 (14.5)52 (23.1)22 (25.6)0.248Combination ATB, n (%)43 (11.5)5 (8.1)21 (9.3)17 (19.8)0.033Overall mortality, n (%)106 (28.4)15 (24.2)65 (28.9)26 (30.2)0.710Abbreviations: AKI, acute kidney injury; ATB, antimicrobial therapy; Cr, creatinine; DIC, disseminated intravascular coagulation; DOI, duration of illness; IE, infective endocarditis; PEN-S, penicillin-susceptible; PEN-I, penicillin-intermediate; PEN-R, penicillin-resistanta Some patients had infections at multiple sitesb Transthoracic echocardiography was performed in 301 patients Table 2Baseline characteristics of 373 S. suis infection casesCharacteristicsTotal(n = 373)PEN-S(n = 62)PEN-I(n = 225)PEN-R(n = 86)P valueDemographics data, n (%) Age ≥ 60 years158 (42.4)28 (45.2)94 (41.8)36 (41.9)0.892 Male262 (70.2)45 (72.6)156(69.3)61 (70.9)0.899 DOI ≥ 7 days129 (34.6)2 (3.2)81 (36.0)46 (53.5)< 0.001 Consume/contact raw pork products60 (16.1)10 (16.1)31 (13.8)19 (22.1)0.202Comorbidities, n (%) Alcoholism62 (16.6)10 (16.1)37 (16.4)15 (17.4)0.980 Diabetes mellitus42 (11.3)10 (16.3)21 (9.3)11 (12.8)0.026Clinical manifestations, n (%) Bacteremia alone96 (25.7)19 (30.6)64 (28.4)13 (15.1)0.030 Complicated bacteremiaa271 (72.4)42 (67.8)157 (69.8)72 (83.7) • Meningitis102 (27.4)32 (51.6)60 (26.7)10 (11.6)< 0.001 • IEb129/3012/44 (4.5)74/178 (41.6)53/79 (67.1)< 0.001 • IEb + Meningitis27/3013/44 (6.8)17/178 (9.6)7/79 (8.9)0.872 • Spondylodiscitis12 (10.9)4 (17.4)6 (8.4)2 (12.5)0.315 No Bacteremia6 (1.6)1 (1.6)4 (1.8)1 (1.2)0.928Complications, n (%) Respiratory failure108 (29.0)17 (27.4)62 (27.7)29 (33.7)0.551 Shock80 (21.4)9 (14.5)52 (23.1)19 (22.1)0.347 AKI (Cr ≥ 2 mg/dL)57 (15.3)8 (12.9)35 (15.6)14 (16.3)0.861 DIC44 (11.8)5 (8.1)30 (13.3)9 (10.5)0.527 Multiorgan dysfunction83 (22.3)9 (14.5)52 (23.1)22 (25.6)0.248Combination ATB, n (%)43 (11.5)5 (8.1)21 (9.3)17 (19.8)0.033Overall mortality, n (%)106 (28.4)15 (24.2)65 (28.9)26 (30.2)0.710Abbreviations: AKI, acute kidney injury; ATB, antimicrobial therapy; Cr, creatinine; DIC, disseminated intravascular coagulation; DOI, duration of illness; IE, infective endocarditis; PEN-S, penicillin-susceptible; PEN-I, penicillin-intermediate; PEN-R, penicillin-resistanta Some patients had infections at multiple sitesb Transthoracic echocardiography was performed in 301 patients Baseline characteristics of 373 S. suis infection cases Consume/contact raw pork products Abbreviations: AKI, acute kidney injury; ATB, antimicrobial therapy; Cr, creatinine; DIC, disseminated intravascular coagulation; DOI, duration of illness; IE, infective endocarditis; PEN-S, penicillin-susceptible; PEN-I, penicillin-intermediate; PEN-R, penicillin-resistant a Some patients had infections at multiple sites b Transthoracic echocardiography was performed in 301 patients The overall case fatality rate was 28.4% (95% CI, 23.8–33.3) and did not differ significantly among the PEN-S, PEN-I, and PEN-R groups (P = 0.710; Table 2). When stratified by clinical manifestation, mortality was 33.3% (95% CI, 25.3–42.2) among patients with IE, 20.6% (95% CI, 13.2–29.7) among patients with meningitis, and 14.8% (95% CI, 0.4–33.7) among patients with concurrent IE and meningitis. The single patient without bacteremia who presented with gangrenous cholecystitis died. Multiorgan dysfunction was the leading cause of death (22.2%). Combination antimicrobial therapy was administered to 43 patients (11.5%). Ampicillin plus ceftriaxone was the most frequently prescribed regimen (15 cases), followed by ceftriaxone plus vancomycin (10 cases) and penicillin plus ceftriaxone (8 cases). Most patients who received combination therapy had IE (26/43, 60.5%). Synergy testing was performed on 23 S. suis isolates with varying susceptibilities to PEN and CRO using the checkerboard microdilution method. The clinical characteristics of the 23 cases are summarized in Additional file 2: Table S2. The AMP–CRO combination demonstrated synergism in 39.1% of isolates (Additional file 3: Table S3), whereas the PEN–CRO combination showed synergism in 21.7% (Additional file 4: Table S4). In contrast, the PEN–GEN, AMP–GEN, and CRO–GEN combinations predominantly exhibited indifferent interactions (Additional file 5: Table S5; Additional file 6: Table S6; Additional file 7: Table S7). No antagonistic interactions were observed (Table 3). Table 3Comparison of checkerboard microdilution assay results among 23 S. suis isolatesSynergyactivity aNo. (%) of isolatesPEN-CROAMP-CROPEN-GENAMP-GENCRO-GENSynergism5 (21.7)9 (39.1)2 (8.7)1 (4.3)1 (4.3)Indifference18 (78.3)14 (60.9)21 (91.3)22 (95.7)22 (95.7)Antagonism00000Abbreviations: AMP, ampicillin; CRO, ceftriaxone; GEN, gentamicin; PEN, penicillina Synergy activity: Synergism was defined as a fractional inhibitory concentration index (FICI) ≤ 0.5; indifference as 0.5 < FICI ≤ 4; and antagonism as FICI > 4 Table 3Comparison of checkerboard microdilution assay results among 23 S. suis isolatesSynergyactivity aNo. (%) of isolatesPEN-CROAMP-CROPEN-GENAMP-GENCRO-GENSynergism5 (21.7)9 (39.1)2 (8.7)1 (4.3)1 (4.3)Indifference18 (78.3)14 (60.9)21 (91.3)22 (95.7)22 (95.7)Antagonism00000Abbreviations: AMP, ampicillin; CRO, ceftriaxone; GEN, gentamicin; PEN, penicillina Synergy activity: Synergism was defined as a fractional inhibitory concentration index (FICI) ≤ 0.5; indifference as 0.5 < FICI ≤ 4; and antagonism as FICI > 4 Comparison of checkerboard microdilution assay results among 23 S. suis isolates Abbreviations: AMP, ampicillin; CRO, ceftriaxone; GEN, gentamicin; PEN, penicillin a Synergy activity: Synergism was defined as a fractional inhibitory concentration index (FICI) ≤ 0.5; indifference as 0.5 < FICI ≤ 4; and antagonism as FICI > 4 Discussion Antimicrobial resistance in S. suis represents an escalating global health concern. In Thailand, S. suis isolates from swine have exhibited some of the highest AMR rates in Asia over the past decade, particularly for penicillin (27%) [25]. In the present study, penicillin resistance among human isolates increased markedly, from 7.1% in 2020 to 57.8% in 2024. This trend is generally consistent with previous reports in both human [2, 8, 25] and swine isolates [26]. Given this increase in resistance, the reliability of antimicrobial susceptibility testing methods warrants careful consideration. Penicillin MICs for 350 isolates were determined using the E-test, which is widely used in routine clinical practice and AMR surveillance. Prior studies have shown good to very good agreement between the E-test and broth microdilution (BMD) for streptococci, with most discrepancies being minor (within one doubling dilution) and occurring near clinical breakpoints [27–30]. Nevertheless, because the E-test provides MIC values on a continuous scale, the classification of isolates with borderline MICs may differ from that obtained by BMD, which remains the reference method. This methodological consideration should be taken into account when interpreting the resistance findings. Beyond resistance trends, we examined whether penicillin susceptibility was associated with distinct clinical presentations. To our knowledge, this is the first study to compare clinical characteristics among patients infected with PEN-S, PEN-I, and PEN-R S. suis strains. We observed that meningitis was more frequently associated with PEN-S strains. Similar findings have been reported previously [2, 3], when meningitis predominated before the widespread emergence of penicillin resistance. However, a clear pathophysiological link between antibiotic susceptibility and clinical presentation has not been established. In contrast, penicillin-nonsusceptible isolates appeared to be more closely associated with infective endocarditis (IE). The increasing incidence of S. suis IE worldwide, including in Thailand [10, 31, 32], has led to recommendations for routine echocardiographic screening in all patients with S. suis bacteremia in the 2023 Duke–International Society for Cardiovascular Infectious Diseases (Duke-ISCVID) criteria [33]. In our study, infection with penicillin-nonsusceptible isolates was significantly associated with an increased risk of clinically apparent IE (P < 0.001). However, residual confounding due to disease severity or referral bias cannot be entirely excluded. In addition to established clinical syndromes, we also observed uncommon disease manifestations and exposure patterns. We report the first documented cases of S. suis causing retropharyngeal abscess and gangrenous cholecystitis. In northeastern Thailand, including Nakhon Ratchasima Province, consumption of undercooked or raw food is common. However, only 16% of patients with S. suis infection reported a history of consuming or handling raw pork products. This discrepancy may reflect unrecognized exposure to raw or undercooked pork or recall bias during history taking. To further contextualize these clinical findings, we examined the distribution of serotypes and clonal complexes among the available isolates. Serotype data were available for only a limited proportion of isolates due to loss of viability after transportation. Nevertheless, the serotype distribution was consistent with previous reports from Thailand, with serotype 2 predominating in human infections, followed by serotype 14 [2]. Infective endocarditis occurred across multiple serotypes. Clonal complex (CC) 1 was the most prevalent, in line with prior studies [2], while other CCs were also detected, reflecting the genetic heterogeneity of S. suis isolates. Multilocus sequence typing was performed in only a small subset of isolates, limiting interpretation. Consequently, the study was underpowered to evaluate associations between clonal background, antimicrobial resistance, and virulence characteristics. Against this microbiological and epidemiological background, the implications for antimicrobial therapy warrant consideration. S. suis infections are primarily treated with β-lactam antibiotics, most commonly penicillin (PEN) and ceftriaxone (CRO) [2, 5, 6]. However, interpretation of antimicrobial susceptibility results and selection of combination therapy remain challenging. Susceptibility breakpoints for S. suis are currently extrapolated from CLSI criteria for viridans group streptococci, despite concerns regarding their clinical applicability to S. suis [7–9, 15, 17, 18]. In the present study, in vitro synergism was observed more frequently with AMP–CRO, followed by PEN–CRO, whereas GEN-containing combinations generally demonstrated indifferent interactions. The activity of β-lactam combinations may reflect complementary mechanisms, including differences in penicillin-binding protein (PBP) affinity and enhanced inhibition through targeting multiple PBPs; however, these mechanisms remain speculative and were not directly investigated. As synergy testing was performed in a limited number of S. suis isolates using the checkerboard method, the findings should be interpreted with caution. Given the two-fold dilution design, reliance on visual endpoint assessment, and absence of time–kill confirmation, small differences in MIC interpretation may shift FICI values, particularly near the 0.5 synergy breakpoint. As ceftriaxone–vancomycin combinations were not evaluated, the clinical relevance of these findings remains uncertain and warrants further investigation. In our cohort, ceftriaxone doses were escalated to 4 g/day in patients with meningitis and in those with intermediate CRO susceptibility or complicated bacteremia, particularly in cases of IE. Because ceftriaxone efficacy is dependent on time above the MIC, higher dosing may enhance antibacterial exposure in severe infections but may also increase the risk of neurotoxicity, including encephalopathy and seizures. Further studies are required to assess the safety and effectiveness of this dosing strategy. Vancomycin is generally considered the preferred alternative for ceftriaxone-resistant isolates; however, its clinical utility may be limited by suboptimal tissue penetration, slower bactericidal activity, the need for therapeutic drug monitoring, and the potential for toxicity. The overall fatality rate was 28.4%, higher than that reported in previous studies [32, 34]. Mortality varied by clinical manifestation, with the highest rate observed among patients with IE. Multiorgan failure was the most common cause of death, consistent with previous reports [18]. This study has several limitations. First, it was conducted at a single referral center in an endemic setting, which may limit generalizability. Second, molecular mechanisms underlying penicillin- and ceftriaxone-nonsusceptibility were not investigated, which constrains interpretation of the observed susceptibility patterns. Finally, in vitro synergy testing was performed in a limited number of S. suis isolates using the checkerboard microdilution method without confirmation by time–kill assays; therefore, these findings should be interpreted with caution. Despite these limitations, this study provides additional clinical, microbiological, and epidemiological data that may contribute to understanding S. suis infections in an endemic setting. Conclusion Penicillin-nonsusceptible S. suis was frequently identified in northeastern Thailand and was more commonly observed among cases of IE, while meningitis was more commonly observed among penicillin-susceptible infections. In vitro synergism was most frequently observed with the ampicillin–ceftriaxone combination, warranting further investigation. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1: Table S1. Penicillin MIC distribution and susceptibility rates among 373 S. suis isolates, 2020–2024. Supplementary Material 1: Table S1. Penicillin MIC distribution and susceptibility rates among 373 S. suis isolates, 2020–2024. Supplementary Material 2: Table S2. Clinical characteristics of 23 culture-confirmed S. suis cases undergoing antimicrobial synergy testing by checkerboard microdilution. Supplementary Material 2: Table S2. Clinical characteristics of 23 culture-confirmed S. suis cases undergoing antimicrobial synergy testing by checkerboard microdilution. Supplementary Material 3: Table S3. Combination MICs and FICI values of ampicillin plus ceftriaxone against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 3: Table S3. Combination MICs and FICI values of ampicillin plus ceftriaxone against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 4: Table S4. Combination MICs and FICI values of penicillin plus ceftriaxone against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 4: Table S4. Combination MICs and FICI values of penicillin plus ceftriaxone against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 5: Table S5. Combination MICs and FICI values of penicillin plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 5: Table S5. Combination MICs and FICI values of penicillin plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 6: Table S6. Combination MICs and FICI values of ampicillin plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 6: Table S6. Combination MICs and FICI values of ampicillin plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 7: Table S7. Combination MICs and FICI values of ceftriaxone plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests. Supplementary Material 7: Table S7. Combination MICs and FICI values of ceftriaxone plus gentamicin against clinically isolated S. suis strains, determined by microbroth dilution checkerboard tests.