Abstract
Background.
Influenza predisposes individuals to bacterial co-infections, which can result in disseminated infection and bacteremia. We describe the epidemiology and outcomes of blood culture co-detections among those hospitalized with influenza over 2 influenza seasons.
Methods.
We sampled individuals of all ages from FluSurv-NET, a US population–based surveillance network of persons hospitalized with laboratory-confirmed influenza, during the 2022–2023 and 2023–2024 seasons. Surveillance staff collected information on bacterial blood cultures within 3 days before or 3 days following admission. We described patient characteristics and in-hospital outcomes, stratified by culture positivity, number of positive cultures, and type of co-detection, using unweighted counts and weighted percentages to account for the complex survey design.
Results.
Overall, 14 316 patients were included, with a median (interquartile range) age of 57 (14–74) years, 53.6% female, 52.0% non-Hispanic White, and 25.6% with ≥4 categories of underlying medical conditions. Of these, 50.8% had ≥1 blood cultures obtained and 5.2% overall had ≥1 bacterial co-detections. Intensive care unit admission occurred for 9.6%, 19.2%, 31.2%, and 47.9% among patients with no blood cultures, negative cultures, 1 co-detection, and >1 co-detection documented, respectively; in-hospital mortality occurred in 1.5%, 3.3%, 9.4%, and 14.5%, respectively. Among patients with positive cultures, 22.1% had Staphylococcus aureus and 7.5% had Streptococcus pneumoniae co-detections; both were associated with severe illness (with 22.4% in-hospital mortality each).
Discussion.
Bacterial co-detections in persons hospitalized with influenza were associated with poor in-hospital outcomes. Efforts to prevent severe influenza and bacterial co-infections, including through vaccination, may reduce substantial morbidity and mortality from influenza.
Keywords: influenza, hospitalization, bacteremia, blood culture, outcomes
BACKGROUND
Influenza is a major cause of morbidity and mortality, with an estimated 120 000–710 000 influenza-associated hospitalizations per year in the United States between 2010 and 2024 [1]. Common complications in children and adults hospitalized with influenza include bacterial pneumonia and sepsis [2, 3]. Influenza infection predisposes individuals to severe bacterial infections through various mechanisms [4]. Physical damage to lung parenchyma and ciliary function can promote bacterial growth and tissue invasion, whereas changes to the local immune cells facilitated by the influenza virus can hamper innate immune responses. A prior meta-analysis of predominately hospitalized patients with laboratory-confirmed influenza found a pooled prevalence of bacterial co-infections in respiratory secretions, sputum, or sterile sites of 23% (range 2% to 65%), with Streptococcus pneumoniae and Staphylococcus aureus most commonly identified [5]. The prevalence and relative contribution of bacterial co-infections in persons with influenza-associated illness may differ by site of detection and change over time; for example, the incidence of invasive pneumococcal disease declined markedly in the United States following widespread vaccination against Streptococcus pneumoniae [6, 7].
Bacterial co-infections in persons hospitalized with influenza illness have been associated with an increased risk of severe clinical outcomes, including admission to an intensive care unit (ICU), respiratory failure, and death, as well as an increased cost of care [8–12]. Postmortem studies of individuals who died following influenza infection have also reported a substantial proportion of deaths associated with bacterial co-infections [13].
Using data from the Centers for Disease Control and Prevention (CDC)-sponsored population-based Influenza Hospitalization Surveillance Network (FluSurv-NET), we sought to characterize recent prevalence of bacterial co-detections from blood cultures among persons of all ages hospitalized with laboratory-confirmed influenza and associated outcomes during 2 recent influenza seasons. Secondary objectives included characterizing the most common bacterial co-detections and examining clinical outcomes stratified by number and type of bacterial co-detections.
METHODS
Setting and Design
FluSurv-NET is a US population–based surveillance network that collects information on laboratory-confirmed influenza-associated hospitalizations among residents of all ages within defined catchment areas [14]. This analysis included FluSurv-NET data from the 2022–2023 and 2023–2024 influenza seasons. FluSurv-NET surveillance areas included in this analysis comprised more than 90 counties or county equivalents across 13 states (California, Colorado, Connecticut, Georgia, Maryland, Michigan, Minnesota, New Mexico, New York, Ohio, Oregon, Tennessee, and Utah) with approximately 30 million residents.
Study Population
FluSurv-NET defined a case as any resident of the FluSurv-NET catchment area hospitalized with laboratory-confirmed influenza (primarily detected through molecular assay or rapid antigen testing at the clinician’s discretion) within 14 days before or during admission from October 1 through April 30 of the 2022–2023 and 2023–2024 influenza seasons. Trained surveillance staff collected detailed baseline clinical information (eg, underlying medical conditions) as well as severe clinical measures (eg, ICU admission, vasopressor use, invasive mechanical ventilation, and in-hospital death) on a stratified random sample of cases using a standardized case report form, as previously described [14]. This analysis excluded: (1) nonsampled cases (in which detailed chart reviews were not completed); (2) cases with incomplete chart reviews; and (3) cases in whom there was record of a blood culture being collected but the culture result was unknown.
Blood Culture Classification
Among sampled cases, information was collected on any blood culture tests that were performed for clinical purposes within 3 days before or 3 days following hospital admission. Bacterial culture results were recorded as positive, negative, or unknown; bacterial type classification was collected for cases with positive culture results with up to 10 bacterial pathogens that could be recorded. If a blood culture was positive for 1 bacterial pathogen, 1 result was recorded. If a blood culture grew multiple different pathogens, each positive pathogen was recorded separately. During the 2022–2023 season, surveillance staff were instructed not to record coagulase-negative staphylococci (CoNS), Bacillus species, or Corynebacterium species unless isolated from 2 separate blood cultures; during the 2023–2024 season, staff were instructed to report any positive result regardless of the number of positive blood cultures (≥1). This analysis did not account for differences in instructions across seasons for CoNS, Bacillus, and Corynebacterium species because multiple positive results in 2023–2024 could not be distinguished from a single positive culture result.
We classified a positive culture as a bacterial “co-detection” as we did not have additional information on whether a positive culture result represented a true invasive co-infection or, in some cases, a blood culture contaminant. Common bacterial co-detections were included as dropdown options in a REDCap data collection tool (Supplementary Table 1). Surveillance staff entered additional positive bacterial blood culture results that did not fit a dropdown option as free text entries.
Two infectious disease physicians further characterized blood culture results (both dropdown selections and free text entries) into bacterial types: Acinetobacter (Acinetobacter baumannii, Acinetobacter species); Enterobacter (including Enterobacter aerogenes, Enterobacter cloacae, Enterobacter species); Enterococcus (Enterococcus faecalis, Enterococcus faecium, Enterococcus species); Escherichia coli; Haemophilus influenzae; Klebsiella (Klebsiella pneumoniae, Klebsiella oxytoca, Klebsiella species); Listeria monocytogenes; Neisseria meningitidis; Proteus mirabilis; Pseudomonas aeruginosa; Serratia marcescens; S. aureus; coagulase-negative staphylococcus; Streptococcus pneumoniae; Streptococcus pyogenes/Group A Streptococcus; Other Streptococcus (other than S. pneumoniae or Streptococcus pyogenes/Group A Streptococcus); and a composite group of “Other” for bacterial isolates entered by free text that did not fit into the previous categories, either because a different species was identified than included above or the blood culture was positive without speciation (eg, gram-positive cocci).
Severe Outcome Measures
We evaluated the following in-hospital measures of severe disease: diagnosis of pneumonia [15]; ICU admission; receipt of invasive mechanical ventilation; receipt of noninvasive positive pressure ventilation (continuous positive airway pressure or bilevel positive airway pressure), vasopressor use; renal replacement therapy; in-hospital death; and median (interquartile range [IQR]) length of hospital admission.
Statistical Analysis
Among the sampled persons hospitalized with influenza, we determined proportions in groups defined by whether blood culture was performed and, among those who had ≥1 blood cultures, whether results were positive or negative for bacterial co-detections. The following groups were specified: No blood cultures obtained (group 1); 1 or more negative blood cultures and no positive blood cultures (group 2); 1 positive blood culture result recorded and any number of negative blood cultures (group 3); and 2 or more positive blood culture results documented and any number of negative blood cultures (group 4).
We described baseline patient demographic and clinical characteristics and severe outcomes stratified by these 4 groups. Among those with 1 or more positive blood cultures (groups 3–4), we further characterized the distribution of bacterial co-detection types. Additionally, we described proportions with severe outcomes by common bacterial co-detection types (where the number of cases with a type of co-detection exceeded 50).
Results were described using unweighted counts and weighted percentages, or median and IQRs, with sample weights and 1000 bootstrap replicate weights applied to account for the complex survey design. Fisher exact or χ2 tests were used to compare proportions, and the Wilcoxon rank-sum test was used to evaluate distribution differences between continuous variables (the Wilcoxon rank-sum test ignored replicate weights). All comparisons were 2-sided, and P values <.05 were considered statistically significant. Analyses were conducted using SAS (version 9.4, Cary, NC).
This activity was determined as nonresearch public health surveillance by CDC and conducted consistent with federal laws (45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. Sect. 241(d); 5 U.S.C. Sect. 552a; 44 U.S.C. Sect. 3501 et seq.). State health departments and academic institutional review boards at participating FluSurv-NET sites reviewed and approved these activities as indicated.
RESULTS
Patient Population
Overall, across the 2022–2023 and 2023–2024 influenza seasons, 43 815 laboratory-confirmed influenza hospitalizations were captured among residents of FluSurv-NET catchment areas (Figure 1). Of these, 29 368 were not sampled for chart review, 39 were excluded because of incomplete chart reviews, and 92 were excluded because blood cultures were ordered but results were unknown. Following exclusions, a total of 14 316 persons hospitalized with influenza met inclusion criteria.
Figure 1.
Exclusion flow diagram. aPercentages weighted to account for the complex sampling design.
Of 14 316 who met inclusion criteria (Figure 1), 49.2% had no blood cultures obtained (group 1), 45.7% had 1 or more negative blood cultures and 0 positive blood cultures (group 2), 3.7% had 1 positive blood culture recorded (group 3), and 1.4% had 2 or more positive blood cultures recorded (group 4). In total, of all included patients (with or without blood cultures obtained), 5.2% were identified as having bacterial co-detection with 1 or more positive blood cultures (groups 3–4).
Patient Characteristics by Group
Across groups 1–4, the median (IQR) age was 57 (14–74) years, 53.6% were female, 52.0% were non-Hispanic White, 25.0% non-Hispanic Black, and 13.0% Hispanic of any race, and a majority of patients had multiple underlying medical conditions including 25.6% with 4 or more categories (Table 1). Overall, 11.8% had 1 or more underlying immunocompromising condition. Patients with no blood cultures (group 1) tended to be younger than those who had blood cultures obtained (groups 2–4) (median age 51 years versus 61 years, respectively; P < .01) and more likely to have no underlying medical conditions (21.7% vs 16.9%, respectively; P < .01). Of patients with blood cultures (groups 2–4), those with 1 or more positive blood culture results (groups 3–4) were slightly older on average (median age 64 vs 61 years, respectively; P < .01) and more likely to have 4 or more underlying condition categories (34.6% vs 26.8%) compared to patients with all negative blood cultures (group 2) (Supplementary Table 2). Further comparisons between patients with no blood cultures performed (group 1) versus those with 1 or more positive blood culture results (groups 3–4) are shown in Supplementary Table 3.
Table 1.
Characteristics of Patients With and Without Bacterial Co-detections Among Persons Hospitalized With Influenza
| Characteristic | Total Hospitalized Sample (Groups 1–4) (N = 14 316) n/weighted % |
No Blood Culture Obtained (Group 1) (n = 7446) n/weighted % |
Blood Cultures Negative (Group 2) (n = 6130) n/weighted % |
1 Positive Blood Culture (Group 3) (n = 530) n/weighted % |
2 or More Positive Blood Cultures (Group 4) (n = 210) n/weighted % |
|---|---|---|---|---|---|
| Season | |||||
| 2022–2023 | 7593 (43.7) | 4258 (47.0) | 3009 (41.1) | 264 (40.1) | 62 (27.4) |
| 2023–2024 | 6723 (56.3) | 3188 (53.0) | 3121 (58.9) | 266 (59.9) | 148 (72.6) |
| Median age (IQR) | 57 (14–74) | 51 (9–72) | 61 (20–75) | 64 (43–75) | 65 (50–74) |
| Age group | |||||
| 0–4 y | 1962 (6.5) | 1161 (7.6) | 765 (5.7) | 26 (2.0) | 10 (2.1) |
| 5–17 y | 1959 (7.0) | 1190 (8.5) | 729 (5.9) | 32 (2.8) | 8 (1.8) |
| 18–49y | 2303 (20.6) | 1301 (21.9) | 869 (19.4) | 100 (19.4) | 33 (16.8) |
| 50–64y | 2465 (21.5) | 1227 (20.3) | 1068 (22.2) | 116 (25.5) | 54 (32.3) |
| 65+ y | 5627 (44.4) | 2567 (41.7) | 2699 (46.8) | 256 (50.3) | 105 (47.0) |
| Sex | |||||
| Male | 6939 (46.4) | 3554 (44.4) | 3011 (48.1) | 259 (48.2) | 115 (54.3) |
| Female | 7377 (53.6) | 3892 (55.6) | 3119 (51.9) | 271 (51.8) | 95 (45.7) |
| Race/ethnicity | |||||
| White, non-Hispanic | 7580 (52.0) | 3862 (50.8) | 3285 (52.6) | 313 (60.0) | 120 (56.4) |
| Black, non-Hispanic | 3163 (25.0) | 1714 (26.3) | 1311 (24.1) | 105 (20.4) | 33 (17.9) |
| Hispanic | 2073 (13.0) | 1082 (12.9) | 901 (13.2) | 58 (10.8) | 32 (15.0) |
| Other/unknown | 1500 (10.0) | 788 (10.1) | 633 (10.1) | 54 (8.8) | 25 (10.7) |
| Number of underlying medical condition categoriesa | |||||
| 0 | 3631 (19.3) | 2140 (21.7) | 1366 (17.4) | 97 (12.4) | 28 (11.6) |
| 1 | 2812 (18.5) | 1499 (18.5) | 1191 (18.6) | 94 (19.0) | 28 (13.2) |
| 2 | 2656 (19.7) | 1403 (19.9) | 1128 (19.5) | 83 (17.3) | 42 (22.8) |
| 3 | 2067 (17.1) | 975 (16.5) | 960 (17.7) | 94 (16.3) | 38 (18.7) |
| 4+ | 3150 (25.6) | 1429 (23.4) | 1485 (26.8) | 162 (35.0) | 74 (33.7) |
| With immunocompromising condition | |||||
| Yes | 1577 (11.8) | 604 (9.4) | 865 (14.3) | 76 (13.8) | 32 (14.2) |
| No/unknown | 12 328 (85.6) | 6544 (86.9) | 5156 (84.2) | 451 (85.7) | 177 (85.7) |
| Missing | 411 (2.5) | 298 (3.7) | 109 (1.5) | 3 (0.4) | 1 (0.2) |
| Received influenza antiviral treatmentb | |||||
| Yes | 11 037 (80.5) | 5363 (75.5) | 5051 (85.3) | 441 (84.2) | 182 (88.6) |
| No/unknown | 3279 (19.5) | 2083 (24.5) | 1079 (14.7) | 89 (15.8) | 28 (11.4) |
| Median (IQR) days from admission to antiviral treatment initiation among treated | 0 (0–1) | 0 (0–1) | 0 (0–1) | 0 (0–1) | 0 (0–1) |
| Number of blood cultures obtained | |||||
| 0 | 7446 (49.2) | 7446 (100.0) | |||
| 1 | 3996 (28.2) | 3730 (58.0) | 266 (45.3) | ||
| 2 | 2434 (19.6) | 2212 (39.4) | 160 (34.1) | 62 (24.8) | |
| 3 | 155 (1.2) | 63 (1.0) | 51 (12.1) | 41 (17.8) | |
| 4 | 212 (1.4) | 113 (1.4) | 43 (7.0) | 56 (32.3) | |
| 5 | 39 (0.3) | 4 (0.0) | 6 (1.0) | 29 (17.2) | |
| 6 | 24 (0.2) | 7 (0.2) | 3 (0.4) | 14 (6.2) | |
| 7 | 4 (0.0) | 1 (0.1) | 3 (0.5) | ||
| 8 | 5 (0.0) | 1 (0.0) | 4 (1.0) | ||
| 9 | 1 (0.0) | 1 (0.2) |
Abbreviation: IQR, interquartile range.
Underlying medical condition categories included asthma, nonasthma chronic lung disease, chronic metabolic disease, diabetes, cardiovascular disease, blood disorder, immunocompromising condition, liver disease, neurologic/neuromuscular disorder, obesity among persons aged ≥2 y, renal disease, and pregnancy.
Including baloxavir, oseltamivir, peramivir, and/or zanamivir initiated up to 14 d before or during admission; oseltamivir was more commonly administered (96.4%) than baloxavir (3.7%) and peramivir (0.2%). No patients received zanamivir.
Bacterial Species Identified
Among 740 patients with one or more positive blood culture results recorded (groups 3–4), 1052 total positives results were recorded. The most common bacterial species isolated included coagulase-negative Staphylococcus species (28.9% of total positives recorded), S. aureus (22.1%), S. pneumoniae (7.5%), and E. coli (6.5%), with 13.9% in the “other” group with a positive blood culture result not fitting within other predefined bacteria types (Table 2).
Table 2.
Blood Culture Results of Hospitalized Adults With Laboratory-confirmed Influenza With ≥1 Bacterial Co-detection by Blood Culture
| Total Among Cases With 1 or More Positive Cultures (Groups 3–4) | 1 Positive Blood Culture (Group 3) | 2 or More Positive Blood Cultures (Group 4) | ||||
|---|---|---|---|---|---|---|
| Bacterial Type | n | Col % (weighted) | n | Col % (weighted) | n | Col % (weighted) |
| Coagulase-negative staphylococcus | 248 | 28.9% | 152 | 35.2% | 96 | 22.1% |
| Staphylococcus aureus | 274 | 22.1% | 100 | 14.8% | 174 | 30.0% |
| “Other”a | 128 | 13.9% | 87 | 18.4% | 41 | 9.1% |
| Streptococcus pneumoniae | 113 | 7.5% | 64 | 8.0% | 49 | 7.0% |
| Escherichia coli | 71 | 6.5% | 28 | 4.9% | 43 | 8.2% |
| Streptococcus pyogenes/Group A streptococcus | 68 | 6.1% | 25 | 4.7% | 43 | 7.5% |
| Other Streptococcus | 67 | 6.0% | 33 | 6.3% | 34 | 5.7% |
| Enterococcus faecalis; Enterococcus faecium; Enterococcus species | 20 | 2.7% | 10 | 2.7% | 10 | 2.8% |
| Klebsiella pneumoniae; Klebsiella oxytoca; Klebsiella species | 19 | 2.2% | 11 | 2.2% | 8 | 2.2% |
| Proteus mirabilis | 10 | 1.3% | 7 | 1.3% | 3 | 1.2% |
| Pseudomonas aeruginosa | 15 | 1.2% | 6 | 0.6% | 9 | 1.8% |
| Enterobacter aerogenes; Enterobacter cloacae; Enterobacter species | 9 | 1.0% | 1 | 0.1% | 8 | 2.1% |
| Haemophilus influenzae | 6 | 0.5% | 5 | 0.7% | 1 | 0.2% |
| Acinetobacter species; Acinetobacter baumannii; Acinetobacter species | 2 | 0.1% | 1 | 0.1% | 1 | 0.1% |
| Neisseria meningitis | 2 | 0.1% | 0 | 0.0% | 2 | 0.1% |
| Listeria monocytogenes | 0 | 0.0% | 0 | 0.0% | 0 | 0.0% |
| Serratia marcescens | 0 | 0.0% | 0 | 0.0% | 0 | 0.0% |
| Total | 1052 | 100.0% | 530 | 100.0% | 522 | 100.0% |
“Other” included positive bacterial culture results that did not fit into other categories included in Supplementary Table 6.
Among 210 patients with 2 or more positive blood culture results recorded (group 4), 522 positive results were recorded overall (Table 2). The most common isolated bacterial species among patients with 2 or more positive blood cultures recorded included S. aureus (174, 30.0%), coagulase-negative staphylococcus species (96, 22.1%), E. coli (43, 8.2%), and S. pneumoniae (49, 7.0%).
Severe Outcomes by Blood Culture Group
Overall, 35.1% of patients who met inclusion criteria (groups 1–4) were diagnosed with pneumonia, 15.3% were admitted to an ICU, 5.3% received invasive mechanical ventilation, and 2.8% died, with a median length of hospital admission of 4 (IQR 2–7) days (Figure 2, Supplementary Table 4). Group 1 (no blood cultures performed) had the lowest percentage of patients experiencing severe outcomes.
Figure 2.
Weighted percentage of patients who experienced severe clinical outcomes by bacterial blood culture group. Abbreviations: BiPAP, bilevel positive airway pressure; CPAP, continuous positive airway pressure; ICU, intensive care unit.
Among patients with 1 or more blood cultures obtained (groups 2–4), those with all negative blood cultures (group 2) compared to ≥1 positive blood culture (groups 3–4 combined) were less likely to have diagnosed pneumonia (46.3% vs 55.8%, P < .01), ICU admission (19.2% vs 35.9%, P < .01), invasive mechanical ventilation support (7.0% vs 17.0%, P < .01), and in-hospital death (3.3% vs 10.8%, P < .01) as well as a shorter length of hospital admission (median 3 vs 6 days, Wilcoxon rank-sum P < .01) (Supplementary Table 2). Among patients that had 1 or more positive blood cultures (groups 3–4), the percentage who experienced severe outcomes was lower in patients who had 1 positive blood culture recorded (group 3) than in patients who had ≥2 positive blood cultures recorded (group 4) (Figure 2, Supplementary Table 4). Among patients with ≥2 positive blood cultures recorded (group 4), 57.9% had a diagnosis of pneumonia, 47.9% were admitted to an ICU, 22.5% received invasive mechanical ventilation, 31.2% had vasopressor use, 14.5% experienced in-hospital death, and median length of hospital admission was 7 (IQR 3–12) days.
Severe Outcomes by Bacterial Type
Outcomes by common bacterial isolate types are shown in Supplementary Table 5. The percentage of patients experiencing severe in-hospital outcomes varied by species, with the highest percentage of patients experiencing in-hospital death were those with codetection with S. aureus or S. pneumoniae (both 22.4%). The proportion who died was lowest among patients with CoNS bacteremia (6.0%) and “other” bacterial species (7.0%) (listed in Supplementary Table 6).
Results by Age Group
The percentage of patients hospitalized with influenza who had one or more blood cultures obtained was lower in pediatric age groups (42.2% for children aged 0–4 years and 40.2% for those aged 5–17 years) compared to adults aged ≥18 years and highest in adults aged ≥65 years (53.9%) (Supplementary Table 7). Adults were more likely than children to have ≥1 positive blood culture. Across all blood culture groups (groups 1–4), children aged 0–17 years were less likely to have severe in-hospital outcomes compared to younger adults (aged 18–64 years) and older adults (aged ≥65 years) (Supplementary Table 4).
DISCUSSION
Bacterial co-detections in the blood were observed in approximately 5% of persons hospitalized with laboratory-confirmed influenza and associated with severe outcomes based on nationally representative US populated-based surveillance. Approximately one half of persons hospitalized with influenza had ≥1 blood culture obtained. Those with negative blood cultures had worse outcomes than those who did not have any blood cultures obtained. The clinical decision to obtain blood cultures, regardless of the result, may be considered a proxy of illness severity. As observed in previous studies [5, 16], among persons hospitalized with influenza with one or more positive blood cultures, S. aureus was frequently co-detected (22.1%), as was S. pneumoniae (7.5%), both common causes of community-acquired pneumonia. Both S. aureus and S. pneumoniae co-detections were associated with severe in-hospital outcomes, with in-hospital mortality exceeding 20% and more than one half being admitted to an ICU. These findings highlight that invasive bacterial infection including bacteremia is a major driver of excess morbidity and mortality in persons hospitalized with influenza. Reducing the risk of severe influenza illness through measures such as seasonal influenza vaccination and vaccination against other vaccine-preventable diseases such as S. pneumoniae may reduce the burden of critical sequelae by preventing illness or attenuating severity [17–19]. Future studies examining risk factors for severe co-infections among persons with influenza may further guide prevention and management strategies such as the role of prompt influenza antiviral therapy.
In this analysis of surveillance data, we defined bacterial blood culture co-detections broadly without using strict criteria to define invasive infections. This was done with the goal of evaluating the epidemiology of bacterial co-detections in the bloodstream and associated outcomes by number and type of co-detection. Of note, many of the bacterial species identified in this study, such as S. aureus and S. pneumoniae, almost always represent true infections when isolated from blood cultures. Common commensals, particularly CoNS species, were also commonly detected and may have represented either culture contamination detections or true invasive infections (eg, in patients with indwelling devices). Patients with CoNS co-detections experienced worse clinical outcomes compared to patients with negative blood cultures, including in-hospital mortality of 6.0% versus 3.3%, respectively. This might be due to increased risk of adverse outcomes due to CoNS pathogenicity, additional co-detections in a subset of these patients, or other risk factors for severe outcomes in patients found to have CoNS bacteremia (eg, implantable devices) [20–23]. Patients with multiple positive bacterial blood culture results recorded generally experienced worse outcomes compared to patients with a single positive bacterial culture recorded. This could be related to difference in the relative distribution of bacterial species among patients with multiple positive cultures, including an overrepresentation of S. aureus, and is consistent with previous studies showing persistent bacteremia to be associated with adverse outcomes [24, 25].
This study has a number of strengths and clinical and public health implications. Importantly, we did not restrict the analysis to only those with positive blood cultures. This provided insights into clinician-driven blood culture testing practices from representative population-based surveillance data of persons hospitalized with influenza during recent seasons and enabled us to compare patient characteristics and clinical outcomes between groups of patients with positive blood culture results and those with negative blood culture results or without testing. The finding of markedly worse clinical outcomes among patients with positive blood cultures compared to patients without positive cultures highlights the importance of preventive strategies, including vaccination against vaccine-preventable diseases that cause severe respiratory illness, and recognizing and providing more intensive clinical management for those hospitalized with influenza with suspected or confirmed co-infections. The large and robust sample of individuals hospitalized with influenza identified from multistate population-based surveillance data also allowed us to stratify findings within clinically relevant subgroups, including stratification of clinician-driven blood culture testing and outcomes within age groups and by common bacterial isolates identified from positive cultures.
This study is also subject to several limitations. This analysis was limited to data obtained through medical chart abstraction using a standardized case report form in a large hospital-based surveillance network for which detailed information on other risk factors for severe outcomes such as vaccination history, clinical management, and other complications such as invasive fungal infections were not collected [26, 27]. Findings should be interpreted in the context of data collection procedures. For example, instructions for documenting positive results for common commensals including CoNS, Bacillus species, and Corynebacterium species, changed between seasons. Furthermore, some types of bacterial detections, such as CoNS species and “other” co-detections, were combined into a single group, whereas individual species may be associated with differential risk for severe outcomes [21, 23]. The short study period of 2 recent influenza seasons precludes the ability to assess long-term temporal changes. However, that is offset by the timeliness of this study in providing a needed update on the prevalence and etiology of bacterial co-detections in persons hospitalized with influenza. Although representing population-based data across 13 states, captured persons hospitalized with influenza in FluSurv-NET surveillance sites may not be representative of the entire US population.
CONCLUSION
Bacterial co-detection in blood culture was found in approximately 5% of persons hospitalized with influenza in a large national population-based network. Outcomes varied by organism, with severe clinical outcomes commonly observed with S. aureus and S. pneumoniae. Efforts to reduce the risk of severe influenza and bacterial co-infections, as well as to promptly recognize and treat these infections, may improve patient outcomes.
Supplementary Material
Supplementary materials are available at Clinical 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.
Acknowledgments.
California: Cristina Curran, Tim Akito Rei, Brenna Hall, Jeremy Roland, Monica Napoles, Madeline Anderson, Allizah Lansang, Justin Ho; CDC: Charisse Cummings, Angelle Kardorff; Colorado: Diane Ankovic, Elizabeth Austin, Violet Bain, Henley Huang; Connecticut: Julia Desiato, Sarah Katsandres, Daewi Kim, Noelle Labozzo, Emily Zmek; Georgia: Allison Roebling, Alyssa Clausen, Annabel Patterson, Asmith Joseph, Chandler Surell, Emily Bacon, Emily Fawcett, Emma Grace Turner, Hope Wilson, Johanna Hernandez, MaCayla Servais, Meghann Cantey, Rayna Ceaser, Sabrina Hendrick, Sydney Hagley-Alexander, Trisha Deshmukh; Maryland: David Blythe, Alicia Brooks, Rachel Park, Larisa Chapa, Claire Summa, Cindy Bryant; Michigan: Ryan Malosh, Taylor Olsabeck, Pauline Stedman, Lauren Leegwater, Liz Urlaub, Sarah Rojewski, Ava Burzycki, Mitchell Magliocco; Minnesota: Alli Johnson, Cynthia Kenyon, Cassy Wilson-Homel, Anna Chon, Karissa Helvig; New Mexico: Daniel M. Sosin, Yomei P. Shaw, Marc Martinez, Murtada Khalifa; New York—Albany: Aleah Abdellatif, Bridget Anderson, Fiona Keating, Suzanne McGuire, Adam Rowe, Jemma Rowlands; New York—Rochester: Christina Felsen, Erin Licherdell, Sophrena Bushey, Katherine St. George, Kevin Popham; Oregon: Victoria Briese, Arilene Novak; Tennessee: Tiffanie Markus, Karen Lieb, Katie Dyer, Terri McMinn, Danielle Ndi, Gail Hughett; Utah: Andrea George, Emma Mendez-Edwards, Hafsa Zahid, Holly Staten, Isabella Reyes, Kristen P. Olsen, Mary Hill, Melanie T. Crossland, Tiana Lui, Katherine Miller, Michelle Vowles.
Financial support.
The Influenza Hospitalization Surveillance Network (FluSurv-NET) is a collaboration of state health departments, academic institutions, and local partners and is funded by the Centers for Disease Control and Prevention (CDC). This work was supported by CDC through an Emerging Infections Program cooperative agreement (grant CK17-1701) and through a Council of State and Territorial Epidemiologists cooperative agreement (grant NU38OT 000297-02-00).
Potential conflicts of interest.
H. K. T. has received CDC research funding. I. A. reports grants from CDC during the conduct of the study, and grants from CDC outside the submitted work. J. M. reports grants from CDC during the conduct of the study. K. Y.-H. reports grants from CDC during the conduct of the study. L. S. W. reports grants from CDC-Emerging Infections Program (EIP) during the conduct of the study. P. A. R. reports institutional funding from EIP cooperative agreement during the conduct of the study. M. L. M. reports institutional funding from EIP cooperative agreement during the conduct of the study. S. K. reports grants from Michigan Department of Health and Human Services during the conduct of the study. V. T. N. reports grants from Michigan Department of Health and Human Services during the conduct of the study. R. L. reports grants from CDC, EIP cooperative agreement during the conduct of the study, and received fee for serving as an Associate Editor for the American Academy of Pediatrics Red Book (Report of the Committee on Infectious Diseases). M. M. reports grants from CDC-EIP during the conduct of the study. C. M. reports grants from CDC during the conduct of the study. M. B. reports grants from CDC-EIP during the conduct of the study. N. E. M. reports grants from Council of State and Territorial Epidemiologists during the conduct of the study. M. S. reports grants from CDC-EIP during the conduct of the study. W. S. reports grants from CDC during the conduct of the study. All other authors report no potential conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Footnotes
Disclaimer. The findings and conclusions of this report are those of the authors and do not necessarily reflect the official position of the Centers for Disease Control and Prevention.
Data availability statement.
The data underlying this article may be shared on reasonable request. Individuals interested in receiving a limited dataset can submit a brief proposal to Dr. Catherine H. Bozio (ise7@cdc.gov) for review and consideration.
References
- 1.Centers for Disease Control and Prevention. About estimated flu burden. Available at: https://www.cdc.gov/flu-burden/php/about/index.html. Accessed 28 June 2025.
- 2.Chow EJ, Rolfes MA, O’Halloran A, et al. Respiratory and nonrespiratory diagnoses associated with influenza in hospitalized adults. JAMA Netw Open 2020; 3:e201323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kalil AC, Thomas PG. Influenza virus-related critical illness: pathophysiology and epidemiology. Crit Care 2019; 23:258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.McCullers JA. The co-pathogenesis of influenza viruses with bacteria in the lung. Nat Rev Microbiol 2014; 12:252–62. [DOI] [PubMed] [Google Scholar]
- 5.Klein YE, Monteforte B, Gupta A, et al. The frequency of influenza and bacterial coinfection: a systematic review and meta-analysis. Influenza Other Respir Viruses 2016; 10:394–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Shah SS, Ratner AJ. Trends in invasive pneumococcal disease-associated hospitalizations. Clin Infect Dis 2006; 42:e1–5. [DOI] [PubMed] [Google Scholar]
- 7.Pilishvili T, Lexau C, Farley MM, et al. Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J Infect Dis 2010; 201:32–41. [DOI] [PubMed] [Google Scholar]
- 8.Bartley PS, Deshpande A, Yu PC, et al. Bacterial coinfection in influenza pneumonia: rates, pathogens, and outcomes. Infect Control Hosp Epidemiol 2022; 43:212–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Martin-Loeches I, Schultz MJ, Vincent JL, et al. Increased incidence of co-infection in critically ill patients with influenza. Intensive Care Med 2017; 43:48–58. [DOI] [PubMed] [Google Scholar]
- 10.Qiao M, Moyes G, Zhu F, Li Y, Wang X. The prevalence of influenza bacterial co-infection and its role in disease severity: a systematic review and meta-analysis. J Glob Health 2023; 13:04063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shah SN, Greenberg AJ, McNulty CM, et al. Bacterial and viral co-infections complicating severe influenza: incidence and impact among 507 U.S. patients, 2013–14. J Clin Virol 2016; 80:12–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu Y, Ling L, Wong SH, et al. Outcomes of respiratory viral-bacterial co-infection in adult hospitalized patients. EClinicalMedicine 2021; 37:100955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Macintyre RC, Chughtai AA, Barnes M, et al. The role of pneumonia and secondary bacterial infection in fatal and serious outcomes of pandemic influenza a(H1N1)pdm09. BMC Infect Dis 2018; 18:637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Naquin A, O’Halloran A, Ujamaa D, et al. Laboratory-confirmed influenza-associated hospitalizations among children and adults—influenza hospitalization surveillance network, United States, 2010–2023. MMWR Surveill Summ 2024; 73:1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tenforde MW, Noah KP, O'Halloran AC, et al. Timing of influenza antiviral therapy and risk of death in adults hospitalized with influenza-associated pneumonia, influenza hospitalization surveillance network (FluSurv-NET), 2012–2019. Clin Infect Dis 2025; 80:461–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Joseph C, Togawa Y, Shindo N. Bacterial and viral infections associated with influenza. Influenza Other Respir Viruses 2013; 7 Suppl 2:105–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Smith AM, Huber VC. The unexpected impact of vaccines on secondary bacterial infections following influenza. Viral Immunol 2018; 31:159–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ferdinands MJ, Thompson GM, Blanton L, Spencer S, Grant L, Fry MA. Does influenza vaccination attenuate the severity of breakthrough infections? A narrative review and recommendations for further research. Vaccine 2021; 39:3678–95. [DOI] [PubMed] [Google Scholar]
- 19.Tessmer A, Welte T, Schmidt-Ott R, et al. Influenza vaccination is associated with reduced severity of community-acquired pneumonia. Eur Respir J 2011; 38:147–53. [DOI] [PubMed] [Google Scholar]
- 20.Martin MA, Pfaller MA, Wenzel RP. Coagulase-negative staphylococcal bacteremia. Mortality and hospital stay. Ann Intern Med 1989; 110:9–16. [DOI] [PubMed] [Google Scholar]
- 21.Becker K, Heilmann C, Peters G. Coagulase-negative staphylococci. Clin Microbiol Rev 2014; 27:870–926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Varisco B, Martínez Pérez-Crespo PM, Retamar-Gentil P, et al. Mortality predictors and definition proposal for complicated coagulase-negative Staphylococcus bacteraemia: a multicentre prospective cohort study. Clin Microbiol Infect 2025; 31:607–15. [DOI] [PubMed] [Google Scholar]
- 23.Keogh S, Larsen EN, Edwards F, et al. Speciation of coagulase-negative staphylococci: a cohort study on clinical relevance and outcomes. Infect Dis Health 2025; 30:307–15. [DOI] [PubMed] [Google Scholar]
- 24.van Hal SJ, Jensen SO, Vaska VL, Espedido BA, Paterson DL, Gosbell IB. Predictors of mortality in Staphylococcus aureus bacteremia. Clin Microbiol Rev 2012; 25:362–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Minejima E, Mai N, Bui N, et al. Defining the breakpoint duration of Staphylococcus aureus bacteremia predictive of poor outcomes. Clin Infect Dis 2020; 70:566–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schauwvlieghe AFAD, Rijnders BJA, Philips N, et al. Invasive aspergillosis in patients admitted to the intensive care unit with severe influenza: a retrospective cohort study. Lancet Respir Med 2018; 6:782–92. [DOI] [PubMed] [Google Scholar]
- 27.Vanderbeke L, Spriet I, Breynaert C, Rijnders BJA, Verweij PE, Wauters J. Invasive pulmonary aspergillosis complicating severe influenza: epidemiology, diagnosis and treatment. Curr Opin Infect Dis 2018; 31:471–80. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article may be shared on reasonable request. Individuals interested in receiving a limited dataset can submit a brief proposal to Dr. Catherine H. Bozio (ise7@cdc.gov) for review and consideration.