«INFLAMMATORY PROTEINS, GENETIC VARIATION, AND ENVIRONMENTAL INFLUENCES ON HEALTH CARE ASSOCIATED INFECTION DEVELOPMENT IN SEPSIS A Dissertation ...»
associated infections. The investigator performed a retrospective 6 week feasibility study in the ICU where this dissertation study subsequently occurred.24 Since cytokine measurements were not clinically available, the relationship of baseline C-reactive protein (CRP) to the development of HAI was examined. All patients admitted to the MICU/CICU for 6 weeks beginning 1/1/08 were screened for inclusion. All subjects with a CRP drawn within 24 hours of ICU admission were included. Baseline demographics, CRP, diagnoses, SIRS, APACHE II, MODS, infections, and invasive devices were recorded and all patient records were reviewed through ICU discharge for the development of HAI. Comparisons were made among those who did and did not develop HAI.
Among 69 patients admitted to the ICU during the study period, 27 patients had CRP levels within the first 24 hours. Among the 27 patients included in this older (62.3 ± 12.9), male (96.3%) heterogeneous population, the mean APACHE II score was 17.8 ± 7.0; 25.9% required vasopressors, 59.6% had suspected or definitive baseline infections, 40.7% required mechanical ventilation, and the median baseline CRP level was 58.4 mg/dL with wide variation (range 3 - 548). Nine (33.3%) patients developed HAI (total of 15 infections) with bacteremia and UTI being most common. There were no significant differences in survival, baseline CRP, APACHE II, SIRS, MODS, or hospital length of stay in those who did and did not develop NI. There was a trend (p = 0.10) towards development of a difference in ICU length of stay (10.4 vs. 4.4) in those who developed new HAI versus those who did not, respectively. Patients with baseline infections were more likely to develop new HAI 8 (88.9%) compared to those who did not present with an infection 1(11.1%), p = 0.04. Although this was a small sample, it suggests increased susceptibility to new HAI in those who were admitted with sepsis.
This study was designed as a prospective observational study to evaluate the effects of baseline systemic inflammation as measured by cytokine levels on the development of HAIs in patients admitted to the ICU with sepsis or suspected sepsis.
Baseline pro-inflammatory cytokine, anti-inflammatory cytokine, and their ratios, along with a common single nucleotide polymorphism for each cytokine tested were examined among patients with sepsis who did and did not develop subsequent HAIs. Approval for human subject’s research was obtained from the Memphis-Veteran’s Affairs Medical Center (VAMC) Institutional Review Board (IRB) and University of Tennessee Health Science Center (UTHSC) IRB, as well as the VAMC’s Research and Development committee.
All patients admitted to the Veterans Affairs Medical Center’s Medical Intensive Care Unit (MICU) were systematically screened for sepsis at admission. A sample size of 78 subjects was required to detect a 30% difference (40% in those with high systemic inflammation versus 10% in those with low systemic inflammation) in development of at least one HAI during ICU stay (up to ICU day 28). This calculation was estimated for patients with high baseline levels of systemic inflammation (pro-inflammatory cytokine level within 4th quartile) versus patients without high levels of systemic inflammation (pro-inflammatory cytokine level not within 4th quartile) with an 80% power given a 3:1 ratio.
The following two conditions were inclusion criteria:
1. Presence of Sepsis within 72 hours of admission to the ICU: Sepsis is defined as systemic inflammatory response syndrome (SIRS) plus an identified or suspected infection. SIRS is defined as at least two of the following: abnormal temperature ( 38°C or 36°C), minimum heart rate ( 90 per minute), elevated respiratory rate ( 20 per minute) or a PaCO2 of 32 mmHg, and an abnormal white blood cell count ( 12,000 or 4,000 cells μL−1) or bands 10%.
2. Age 18 or older.
The following four conditions were exclusion criteria:
1. Receiving Immunosuppressive agents such as chemotherapy or greater than 1 mg/kg of prednisone or equivalent dose within the past 3 months.
2. Other Immunosuppression: AIDS.
3. Liver Failure defined as Child Pugh category C or higher.
4. Lack of informed consent.
This section includes study related procedures.
Summary of Procedures A summary of all study procedures over time is included in Table 3-1. Procedures included screening, informed consent, data collection, specimen collection, and specimen analysis.
Screening Efforts were made to screen all patients for inclusion and exclusion criteria who were admitted to the medical intensive care unit (MICU). Rounds were systematically made in the MICU and the principle investigator interacted with the MICU medial team, nursing staff, and reviewed records to determine if patients met eligibility. A waiver of informed consent was obtained from the IRB to allow for screening of patient records for eligibility. Patient records included their electronic chart, nursing flowsheets, x-ray films, and other pertinent reports needed to determine eligibility. The principle investigator completed training in human subject’s protection prior to initiation of this research.
Subjects who met inclusion criteria and had no exclusion criteria were approached for informed consent. When subjects were too ill to provide their own informed consent, then a legally authorized representative/surrogate decision maker was sought for informed consent. All legally authorized representatives (LAR) who provided informed consent met requirement established by VAMC-IRB (see Appendix D, Informed Consent). The LAR designation used for this study was adopted from Tennessee State Law; however, a new designation has recently been adopted based on the latest release of the Veteran’s Affairs Research Handbook which further limits LAR in regards to HIPPA consent for research.
Table 3-1. Study Procedures.
Note: ICU denotes intensive care unit. SIRS denotes systemic inflammatory response score.
As of March 2011, the only authorized LAR will include only individuals appointed under the Durable Power of Attorney for Health Care or a court-appointed guardian.
Data Collection Baseline demographic variables were collected at ICU admission including age, race, and sex. Baseline sepsis (SIRS plus infectious cause) were collected in addition to descriptive variables surrounding baseline sepsis including the presence of organ failure (SOFA and MODS) and severity of illness (APACHE II).
Several predisposing factors were collected. Baseline plasma IL-6 and IL-10 levels were measured as well as their associated SNPS ra1800795 and rs1800896, respectively. Potential environmental exposures were also collected including invasive devices, nurse-patient ratios 2:1 for any 4 hour period each day, and administration of blood products. The number and type of invasive device was recorded each day and the cumulative invasive devise score was recorded at the time of HAI and ICU discharge.
Duration of ICU stay was collected.
Patients were followed through their ICU stay (or up to 28 days) for the presence of HAI based on CDC guidelines. Descriptive variables surrounding the first HAI were recorded including SIRS and organ failure (SOFA and MODS).
In addition to the primary variables a number of other variables were collected to better characterize the sample including vital signs, SIRS score, clinical triggers, all cultures, standard of care laboratory tests, standard of care CXRs, antibiotic use, use of steroids, and HAI preventive strategies.
Data were collected by the principle investigator using standardized forms which were updated during the data collection process (see Appendix E). Information about adverse events (primarily deaths) occurring during the study period were collected and reported to the IRB. The occurrence of life threatening arrhythmias and respiratory and cardiac arrests were also recorded. Data collection after ICU discharge was limited to 48 hours after ICU discharge for assessment of HAIs that could be attributed to ICU stay.
Reliability and Validity of Common ICU Measures
Vital signs are routinely recorded hourly in the ICU unless otherwise indicated.
The values recorded by the critical care nurses during the course of standard patient care were reviewed. There was some variability in these measures and it was observed that respiratory rate was sometimes collected from the EKG respiratory lead (where it is captured and retained by the monitor) rather that from the ventilator display. There were instances of respiratory rates recorded that were lower than the ventilator set rate. When this occurred, the set ventilator rate was recorded as the minimum respiratory rate rather than the rate recorded on the flowsheet. Calculation of APACHE II requires the highest or lowest values within the first 24 hours, along with the worst lab values in this time period. The MICU uses bedside monitors with automated blood pressure cuffs which will provide for consistency in blood pressure readings. Arterial lines and central lines were routinely calibrated and maintained by bedside nurse per hospital policy. Standard quality control procedures in the hospital’s laboratory include calibration of equipment at least every 8 hours and as needed. During the course of this study, Exergen’s Temporal Scanner TAT-5000 Temporal Artery Thermometer became routine use in the MICU.
There were no pilot studies to test the reliability or consistency of this method and it may have introduced some bias in temperature measurement, and therefore SIRS scores.
Participant records were examined prior to data entry to assure that data collection was complete. Clinical data were entered into a FilemakerPro® database.
Prior to data analysis, data was exported to Excel, and all data was reviewed and cross checked for accuracy. Corrections were made to provide the most accurate dataset possible. Laboratory data (cytokine measures) were imported directly into Excel from the Luminex machine. Data from the LightCycler® 480 Instrument were also imported to Excel for data analysis. Tests that did not meet specified conditions were repeated, and excel spreadsheets were updated to reflect additional results.
This section provides detail for study measurements.
Health Care Associated Infections A HAIs were defined as any infection occurring after day 3 through ICU discharge (or day 28 in participants with a prolonged ICU stay), excluding recurrent positive baseline infections. All participants were monitored with equal diligence for the development of HAIs daily through ICU discharge (or day 28 in participants with a prolonged ICU stay). Participants who were discharged from the ICU prior to day 28 were followed for 48 hours in the hospital. All infections that occur up to 48 hours after ICU discharge were also considered ICU related infections, just as infections occurring within 48 hours of hospital admission are considered community acquired. If a participant was readmitted to the ICU within 48 hours of ICU discharge, the patient was followed as though their ICU stay had been continuous.
All culture results during the study period were recorded. Developing an HAI during ICU stay was the primary outcome measure in this study. Each patient was classified as either having or not having a definitive HAI during ICU stay by culture criteria as defined below. Furthermore, the timing to development of first HAI was also recorded. The specific type of HAI and causative organism were recorded. Borderline HAIs were adjudicated by pulmonary and critical care consultation. HAIs were recorded in such a manner to allow for future incidence density calculations. The number and type of devices were recorded daily for each subject.
This study was implemented using the most recent HAI definitions by CDC Guidelines at that time.111 In November 2008, new HAIs guidelines replaced existing guidelines.112 These new guidelines implemented minor changes in HAI definitions but required that the new term “heath care associated infections” replace the prior term “nosocomial infections”. Additionally, the names of common HAI have been updated to reflect the new guidelines. The operational definition for each type of infection did not change. Health care associated infections were defined by CDC Guidelines111 as follows, and terms were revised to meet current guidelines:112 Bloodstream infection (formerly bacteremia) was defined by the presence of • a microorganism cultured from blood which is not related to another site of infection. At least one blood culture is required to be positive, and when the identified organism is a potential skin contaminant such as coagulase-negative staphylococcus, two or more blood cultures must be positive.
Pneumonia was clinically defined as the presence of fever ( 38◦C), • leukopenia ( 4,000 WBC/mm3) or leukocytosis ( 12,000 WBC/mm3), and new or worsening infiltrate on CXR (or the presence of consolidation or cavitation), and at least two of the following: (1) new onset of purulent sputum, or change in sputum of change increased respiratory secretions, or increased suctioning requirement, (2) New onset of worsening cough, or dyspnea, or tachypnea, (3) Rales or bronchial breath sounds, or (4) worsening gas exchange, increased oxygen requirements, or increased ventilation demand. Pneumonia was also definitively defined by positive bronchial alveolar lavage (BAL) with at least 104 cfu/ml.
Urinary tract infection (UTI) can be either symptomatic or asymptomatic.
• Symptomatic UTI is defined as a positive urine culture with 105 microorganisms/ml and one of the following: fever ( 38◦C); frequency;