The Prospective Study Of The Risk Factors Related To Acute Post-stroke Infection And Its Influences On Short-term Prognosis In Acute Stroke

Background: Acute post-stroke infection defined as a variety of infections occurred after 48 hours of acute stroke. It is one of the most common and important complications in acute stroke. Studies showed that more than 20% of stroke patients suffer from infections, and most infections occurred within 7 days after stroke. Acute post-stroke infection is the first cause of death after 1 day of acute stroke. However, now the detailed data about acute post-stroke infections are rare, and also comprehensive and systematic exploration for risk factors related to acute post-stroke infections. In addition, whether the infection is an independent risk factor leading stroke patients to poor prognosis? It is still of many controversies. Therefore, we need to investigate the status of acute post-stroke infections in patients with acute stroke, and confirm the risk factors related to acute post-stroke infections and its influences on short-term prognosis in acute stroke, which will contribute to early identify acute stroke patients with high risk of infections, and take effective prevention measures, finally reduce the incidence of infection and the mortality of stroke.Objective: To investigate the acute post-stroke infections rate in patients with acute stroke, the distribution of infections site, the major pathogens of infections and their antimicrobial sensitivity; To analyze the risk factors related to acute post-stroke infections, seek for the controllable factors which can be intervened in clinic, and facilitate the prevention of acute post-stroke infections; To observe the influences of acute post-stroke infections on short-term prognosis in acute stroke, and further confirm the relationships of acute post-stroke infections with neurological impairment, disability, activities of daily living(ADL), malnutrition, the incidence of other stroke complications, the length of stay(LOS) and the treatment costs; To lay the theoretical basis for strengthening the basic idea of management of acute post-stroke infections and establishing the operation standards to control acute post-stroke infections.Methods: Prospective observational study (cohort study) method was used. We developed "acute post-stroke infections questionnaire", and used it to investigate and observe the inpatients with acute stroke (onset of≤2 weeks) in our hospital from January 1, 2008 to December 31, 2009. Acute post-stroke infection defined as a variety of infections occurred within 48 hours to 2 weeks after acute stroke onset, and specific diagnostic criteria accord to The People's Republic of China Ministry of Health (2001), No. 2 document "diagnostic criteria of nosocomial infection (for trial implementation)." (1) First, we investigated the general demographic items (including age, gender) in all acute stroke patients selected; In the mean time, we investigated the incidence and site of acute post-stroke infections, identified the pathogens of acute post-stroke infections, detected the drug sensitivity of pathogenic bacteria using Kirby-Bauer disc agar diffusion method. Using x±s express measurement data, and using rate(%) and(or) constituent ratio(%) express count data. (2) Referencing to the relevant professional knowledge and the past research literature, we selected the relevant factors which may affect acute post-stroke infections including age, sex, stroke type, stroke frequency, diabetes, chronic respiratory diseases, cancer, anemia, baseline fasting blood glucose (FPG), baseline nutritional status, invasive examination and treatment, Proton Pump Inhibitors(PPIs), corticosteroids, early rehabilitation, posture, consciousness (Glasgow Coma Scale, GCS), neurological deficit level (National Institute of Health Stroke Scale, NIHSS), ADL(Barthel Index, BI), the degree of disability (modified Rankin Scale, mRS) and swallowing function. We investigated the baseline (at admission )factors mentioned above in all acute stroke patients selected. On the one hand, we did single-factor analysis to compare the difference of these factors between infection group and non-infection group, usingχ2-test dealt with count data, using t-test dealt with measurement data, to select the risk factors related to acute post-stroke infections. On the second hand, according to single-factor analysis results and relevant professional knowledge, fitting multi-factor non-conditional Logistic regression model, adjusting the relevant confounded factors, we found the independent risk factors affecting acute post-stroke infections. Using odds ratios (OR) and 95% confidence interval (95% CI) express contact strength between these risk factors and acute stroke-post infections (P<0.05 ). (3)At admission (baseline) and 21d (21±3d)after admission, we investigated indicators of prognosis including neurological deficits (NIHSS), disability (mRS), ADL(BI), nutritional status, other complications, mortality rate, LOS and treatment cost (total cost of hospitalization, drug costs), and compared the difference of these indicators between infection group and non-infection group at admission and 21d. Usingχ2-test, Mcnemar-test or the Fisher exact probability method dealt with two sets of results of count data; using t-test dealt with two sets of results of measurement data in line with normal or near normal distribution; Using rank-sum-test dealt with two sets of results of measurement data is not consistent with the normal distribution (P<0.05). All data were collected by trained clinicians. We conducted quality control and reliability testing of questionnaire, encoded the questionnaire, inputed it to databases tiwce. After proof-reading, we conducted statistical analysis using SPSS for Windows Ver.17.0 statistical software.Results: (1) The results of survey and etiology analysis of acute post-stroke infections:①The incidence of acute post-stroke infections was 18.5%; The site of infection followed by respiratory system, urinary system, multiple sites, and other parts; Incidence of acute post-stroke pneumonia was 11.0%.②Cultured pathogens: Gram-negative bacteria accounted for 67.5%, mainly Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa; Gram-positive bacteria accounted for 25.3%, mainly Staphylococcus aureus; Fungi accounted for 7.2%.③Antimicrobial agents which gram-negative bacilli sensitive to were mainly imipenem, amikacin, netilmicin and tobramycin; Antimicrobial agents which Staphylococcus aureus sensitive to were mainly nitrofurantoin demethylvancomycin and teicoplanin; Producing extra-spectrum beta lactamases (ESBLs) rate of Escherichia coli and Klebsiella pneumoniae were 50.0% and 46.2%, respectively; Staphylococcus aureus strains were all methicillin-resistant Staphylococcus aureus (MRSA). (2)The risk factors of acute post-stroke infections in multi-factor non-conditional Logistic regression analysis: age≥80 years, hemorrhagic stroke, stay in bed, early use of PPIs after stroke, baseline GCS≤8 points, baseline NIHSS≥15 points, baseline BI <50 points, these factors are independent risk factors of acute post-stroke infections (all P <0.05). (3) The influences of acute post-stroke infections on short-term prognosis in acute stroke:①NIHSS improvement is less in infection group than non-infection group, but the difference was not statistically significant[- (1.25±1.39) vs - (1.35±2.55), P = 0.509]; The mean gap scores of MRS, BI between the day of admission and the 21 day after admission were significantaly less in infection group than in non-infection group[- (0.20±0.50) vs - (0.68±0.83), P <0.01; 5.39±8.27 vs 15.29±20.06, P <0.01].②The incidence of malnutrition of infection group was significantly higher at the 21 day after admission than at the day of admission (51.7% vs 31.5%, P <0.01); The incidence of malnutrition of non-infection group was not significantly higher at the 21 day after admission than at the day of admission (4.2% vs of 3.2%, P= 0.388).③The incidence of other complications is higher in infection group than in non-infection group(67.3% vs 16.2%, P<0.01); The mortality was also higher in infection group than in non-infection group(11.9% vs 2.3%, P<0.01).④Patients had shorter LOS(22±8d vs 35±20d, P < 0.01), less total costs of hospitalization(14180±3518$ vs 22939±12966$, P<0.01) and drug costs (6399±2226$ vs 10388±7012$, P<0.01) in non-infection group than in infection group.Conclusions: (1) The incidence of acute post-stroke infections was high (18.5%), mainly in respiratory and urinary systems, multi-site co-infection was common, the incidence of acute post-stroke pneumonia was 11.0%. (2)The pathogens of acute post-stroke infections were mostly opportunistic bacteria, the most of them were Gram-negative bacteria, mainly Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa; The most of Gram-positive bacteria were Staphylococcus aureus, latter were all MRSA; Fungi accounted for 7.2%. (3) Gram-negative bacilli were generally sensitive to the second-generation of aminoglycoside antibiotics, the third-generation of semi-synthetic aminoglycoside antibiotics and carbapenem; MRSA were sensitive to nitrofuran antibacterial drugs and vancomycin. (4) There were many independent risk factors of acute post-stroke infections including age≥80 years, hemorrhagic stroke, stay in bed, early use of PPIs after stroke, baseline GCS≤8 points, baseline NIHSS≥15 points and baseline BI < 50 points. (5) Acute post-stroke infection may lead to increased acute stroke mortality rate, badness of short-term clinical prognosis, prolonged hospitalization and increased treatment costs. (6) To pay more attention to acute post-stroke infections and which has a positive effect to improve the short-term clinical prognosis and the ability of patients returning to society, and to reduce acute stroke mortality rate. Above all, it will contribute to rational and full utilization of limited medical resources of China.