Prevention of device-associated healthcare-associated infections (DA–HAIs), such as central line-associated bloodstream infection (CLABSI) and ventilator-associated pneumonia (VAP), in neonatal intensive care units (NICUs) continues to receive high priority. Surveillance of the occurrence of DA–HAIs is important for the quality improvement of infection control practices at individual institutions. However, no comprehensive benchmark data about the occurrence of DA–HAIs are available for NICUs in Japan. The present study assessed the occurrence of DA–HAIs and investigated the measures necessary for the improvement of overall surveillance in NICUs in Japan. A retrospective cohort DA–HAI surveillance study was conducted in neonates with birth weight of ≤1500 g who were admitted to Level II/III NICUs between January 2011 and December 2012. The diagnosis of DA–HAIs was based on the criteria of the Center for Disease Control and Prevention National Healthcare Safety Network (CDC/NHSN). This study investigated laboratory-confirmed bloodstream infection (LCBI) and clinical sepsis (CSEP) reported as CLABSI. The obtained data were compared with the 2011 CDC/NHSN data summary for Level II/III NICUs. This study included 9 of the 15 NICUs involved in the study of the “Neonatal Infection Control and Prevention Searching Group” in Japan. Ten cases of LCBI, 17 cases of CSEP, and 40 cases of VAP were reported during the 67,788 patient-days of the study period. The rates of LCBI for the birth weight categories of ≤750 g, 751–1000 g, and 1001–1500 g were 0.7, 0.7, and 0.5 per 1,000 device-days, respectively. These rates were all within the standard ranges of the NHSN data. CSEP accounted for 63.0% of the cases of CLABSI. The rates of VAP for the birth weight categories of ≤750 g, 751–1000 g, and 1001–1500 g were 3.3, 5.0, and 3.6, respectively. These rates were all higher than the 75th percentile level of the NHSN data, and varied greatly among the surveyed NICUs. CSEP accounted for a large proportion of CLABSI cases in this study, suggesting that considerable efforts should be made for the accurate diagnosis of LCBI. Furthermore, to achieve additional improvement of infection control practices in NICUs in Japan, large prospective surveillance studies and analysis of the risk factors for each DA–HAI are necessary.
White coats and stethoscopes are known sources of transmission of multidrug-resistant organisms. However, neither the frequency of white coat changes nor that of stethoscope disinfection has been investigated in Japan. A multicenter surveillance study was performed of the frequencies of white coat changes, stethoscope diaphragm disinfection, and the types of white coats used among full-time physicians working at four hospitals in Japan from July to August 2013. The surveillance sheets of 308 physicians (98% of total physicians) were analyzed. The frequency of white coat changes was once a week in half of the physicians (48%) and once daily in 23 physicians (7.5%). In total, 162 physicians (53%) wiped the stethoscope diaphragm during clinical practice, but only 37 physicians wiped the stethoscope diaphragm after examining each patient. Multivariate regression analysis showed that the number of years of clinical experience (≥10 years) was an independent factor associated with a lower frequency of white coat changes (p=0.04), and that male physicians wiped the stethoscope diaphragm less frequently than female physicians (p=0.01). The type of hospital was not associated with the frequency of either white coat changes or stethoscope diaphragm disinfection. It is important for clinicians to change their white coat every day and wipe the stethoscope diaphragm after examining each patient to prevent multidrug-resistant organism transmission. The current surveillance study revealed that most hospital physicians did not perform appropriate white coat changes or stethoscope diaphragm disinfection. Further investigation is required to clarify the factors involved.
Healthcare-associated infections are mainly transmitted by the hands of healthcare workers, so hand hygiene is the most basic and important precaution to prevent infection. The efficacy of hand hygiene depends on both following the appropriate procedure and correct timing. In our hospital, the compliance rate with hand hygiene of our nursing staff has been rising every year as a result of hand hygiene surveillance by the direct observation method conducted since 2010. However, the outbreak of methicillin-resistant Staphylococcus aureus (MRSA) in our ICU in 2012 resulted in identification of various issues with the timing of putting on and taking off gloves in addition to the timing of hand hygiene, so that we initiated new surveillance of the timing issue and the results were given to our nursing staff. Subsequently, the mean compliance rate of timing showed a nonsignificant increase from 90% to 96% (p=0.096), but the detection rate of MRSA significantly decreased from 4.56/1000 patient days to 1.8/1000 patient days (p=0.002). Incidents of MRSA outbreak also diminished. These findings suggest that repeated surveillance by direct observation method, provision of effective education including surveillance results and improvement plans in person and by poster presentation, and improvement of nursing procedures such as environmental arrangements and nursing workflow including cooperation between staff members are effective to increase and maintain the compliance of timing of hand hygiene and putting on and taking off gloves, with consequent reduced detection rate of MRSA.
Infection control team (ICT) monitoring is considered an active approach to manage infection in hospitals. The ICT was formed in August 2004 at our hospital, and supportive measures for the appropriate use of antimicrobial agents were gradually implemented. One of these measures was the ICT antimicrobial agent monitoring program (round) initiated in May 2010. The effectiveness of the round was studied in 265 of 387 cases between May 2010 and March 2013 without death or discharge at less than 3 days after the round without laboratory data or chart. These cases were divided into change, stop, and continuance groups. Changes in white blood cell (WBC) count, C–reactive protein (CRP), and body temperature (BT) were analyzed between and after the rounds. CRP in the stop group showed a statistically significant reduction, and WBC count and BT in the stop group exhibited declining trends. WBC count, CRP, and BT in the change and continuance groups showed statistically significant reductions. Although there were deteriorating cases of microbial substitution or underlying disease, recommended treatment by the ICT was assumed to be effective because of the absence of infection ingravescence. As a result, the round is predictably effective for appropriate use of antimicrobial agents and infectious disease therapy by controlling prolonged administration: furthermore, the round can decide on changes in appropriate dose or de-escalation. Based on this finding, we recommend interventions on the appropriate use of high-quality antimicrobial agents to continuously support effective treatment for infectious diseases.
The association of antibiotics with Clostridium difficile infection (CDI) has been extensively described. However, few reports have described cases in Japanese patients. Therefore, the association between CDI and each class of antibiotics was examined in our hospital. The medical records of all patients were examined in whom C. difficile toxin A (CD toxin A) testing for suspected CDI was performed during the period from July 2005 to December 2007. The risk of developing CDI was evaluated by comparing the groups of CD toxin A–positive patients and negative patients as the control group. Onset of diarrhea during the period was evaluated by a survey of the medical records of 269 cases. Diarrhea occurred in 56 patients who received antibiotics and in 3 patients who did not receive antibiotics in the CD toxin A–positive group, but in 169 cases and 41 cases, respectively, in the control group. Comparison of the two groups found that the classes of antibiotics which increased CDI significantly were fluoroquinolones (adjusted odds ratio [AOR], 9.0 [95% confidence interval, 2.7–29.9]), second-generation cephalosporins (AOR, 7.2 [95% confidence interval, 2.4–22.1]), and third-generation cephalosporins (AOR, 4.1 [95% confidence interval, 1.4–11.8]). A multi-center study and a survey of the relevance of CDI and specific antibiotics are needed.
An outbreak of adenovirus-associated epidemic keratoconjunctivitis (EKC) occurred in the day-care center in our hospital, and the center was closed for a week in June 2011. This study documents the effect of intervention by the infection control team (ICT) on an outbreak of EKC in a hospital day-care center. This outbreak was caused by delayed intervention of the ICT and inadequate contact infection control. The recommended improved infection control measures were (1) early intervention of the ICT in the hospital day-care center, (2) early definition of cohorts of infected child patients followed by diagnosis and return home, and (3) enhanced adherence to contact infection control and environmental disinfection measures. After implementation of the control measures, the infection rate of EKC was reduced, but not significantly (26% in 2011, 8% in 2012; p=0.093). This study suggests that early definition of cohorts of child patients followed by diagnosis and return home, and enhanced adherence to contact infection control and environmental disinfection measures may be essential to prevent an outbreak of EKC in a hospital day-care center.