Preventing Postoperative Healthcare Associated Infections: The Experience of Two European Cardiac Surgery Intensive Care Units

The phenomenon of healthcare–associated infections (HAIs), formerly called nosocomial infections (NIs) [1], has been well recognized in the literature. In the cardiovascular ICUs, HAIs may result in significant patient morbidity and mortality, prolonged duration of hospital stay, and additional costs due to diagnostic and therapeutic interventions [2]. For this reason, HAIs are considered an indicator of the quality of patient care, an adverse event, and a patient safety issue. The use of illness severity scores is very common in Intensive Care Units (ICUs) in order to predict outcome, differentiate disease severity and assess resource use [3]. The Simplified Acute Physiology Score II (SAPS II) [4] is a very commonly used score in Europe to compare a critically ill patient’s severity and to evaluate clinical course and outcome [5].


Background
The phenomenon of healthcare-associated infections (HAIs), formerly called nosocomial infections (NIs) [1], has been well recognized in the literature. In the cardiovascular ICUs, HAIs may result in significant patient morbidity and mortality, prolonged duration of hospital stay, and additional costs due to diagnostic and therapeutic interventions [2]. For this reason, HAIs are considered an indicator of the quality of patient care, an adverse event, and a patient safety issue. The use of illness severity scores is very common in Intensive Care Units (ICUs) in order to predict outcome, differentiate disease severity and assess resource use [3]. The Simplified Acute Physiology Score II (SAPS II) [4] is a very commonly used score in Europe to compare a critically ill patient's severity and to evaluate clinical course and outcome [5].
Strategies of surveillance implementation [6,7] may result in decreasing HAI rates.
Following the successful of the US system of National Nosocomial Infection Surveillance Scheme (CDC/NHSN), a national network [8] for the surveillance of NIs has been developed in Europe since the early '90s (HELICS/IPSE).

Summary Statement
Why is this research or review needed?
• The phenomenon of healthcare-associated infections (HAIs), in the cardiovascular ICUs may result in significant patient morbidity and mortality, prolonged duration of hospital stay, and additional costs due to diagnostic and therapeutic interventions • HAIs are considered an indicator of the quality of patient care, an adverse event, and a patient safety issue.
What are the key findings?
• Nursing activities are more effective in reducing the risk of HAIs when performed in accordance with a prevention protocol.
• Nursing assistants help reducing the time spent in the everyday direct patient care.
• Excessive nursing workloads may negatively influence the time spent in HAIs surveillance activities.
How should the findings be used to influence policy/ practice/research/education?
• The presence of nursing assistants in ICUs should be encouraged.
• Nurses should be allowed to spend more time in NI

Observational Study
Preventing Postoperative Healthcare Associated Infections: The Experience of Two European Cardiac Surgery Intensive Care Units to poor adherence to basic infection control measures (e.g., hand hygiene), and higher rates of adverse outcomes (e.g., HAIs) [9]. The analysis of the nursing staff workloads is possible through specific indicators, such as patients' illness severity scores 4 and nursing activities scores [10].

Aim
The aim of the study was to compare the time spent by the nursing staff in surveillance activities during the first shift -the most intense in terms of workloads -in two South-European Cardiac Surgery ICUs.

Methods
We compared the two ICUs analyzing two aspects: the severity of disease and the nursing staff workloads. The first variable was assessed using the Simplified Acute Physiology Score II (SAPS II), and the nursing staff workloads were analyzed using the Nine Equivalents of Nursing Manpower Score (NEMS). We informed the eligible patients that we were carrying out a quality assurance program, and that we would collect data about their health conditions.
We also asked permission to observe the nursing staff activities to both Hospital directions.
The observers' researchers were trained in the use of these tools through a brief didactic course.

Simplified Acute Physiology Score II
This tool was developed and validated in France in 1984 in order to measure the severity of disease and to predict the risk of death in ICU patients aged 15 or more. Like the APACHE scores [12], SAPS was calculated from the worst values obtained during the first 24 hours after admission to the ICU. In 1993, Le Gall and colleagues [4] used logistic regression analysis to develop SAPS II, which includes 17 variables: 12 physiological variables, age, type of admission, and 3 variables related to underlying disease. The scores may vary from 0 to 163, and the predicted mortality from 0% to 100%. The score is calculated only once during the stay. The SAPS II score was validated using data from consecutive admissions to 137 ICUs in 12 countries. Each score corresponds to a recommended nurse to patients ratios.

Nine Equivalents of Nursing Manpower Score
This tool was derived from the TISS-28 in order to create a simpler system, easier to use [10]. EMS can be used to evaluate the efficacy of nursing workload use at the ICU, and classify ICUs based on the nurses' activities provided [3]. The over mentioned activities are divided into nine categories: basic monitoring, intravenous medication, mechanical ventilatory support, supplementary ventilatory care, single vaso-active medication, multiple vaso-active medication, dialysis techniques and specific interventions inside and outside the ICU. Each variable is weighted with a score, to a maximum of 56. Each score corresponds to a recommended nurse to patients ratios.

Data collection and analysis
Patients: we considered eligible all the patients who underwent cardiac surgery during the period of observation and accepted to enter the study. The sample was composed of 60 patients per ward, consecutively recruited.
Setting and period of observation: We collected the data at the Unidad de CuidadosIntesivos of the Hospital "Rio Hortega", Valladolid (Spain), and the Post-Operative Cardiac Surgery Unit of the University Hospital, Ancona (Italy) in the period 1st February -30th June 2010. We observed the nursing workloads during the first shift (from 7 a.m to 2 p.m.), and we collected data about activities of NI surveillance performed by nurses during this time.

Statistics:
The inter-rater reliability (IRR) of the measurements performed through the SAPS II and the NEMS tools was assessed with the IRR statistics Quickcalcs -Grapf Pad SW© [11]. The variables "nursing workload" and "NI surveillance" were compared using the Chi square test statistic.

Results
The mean scores observed in the Italian and Spanish ICUs are similar for both the SAPS II (40.83 and 41, respectively) and the NEMS tool (49 and 50, respectively). The assessment of the inter-rater reliability (IRR) of data, using QuickCalcs -Grapf Pad SW © shows a Kappa value of 0.523 (95% C.I. 0.437 -0.610, SE = 0.044) (Tables 1  and 2).
We also observed the NI surveillance activities performed by the nursing staff scheduled during the first shift in both ICUs (Table 3). In the Spanish ICU, these activities were performed in accordance with the government project "Bacteriemia zero" [13], which is part of a protocol of global strategy of HAI prevention in the ICUs supported by the Joint Commission.
The distribution of nurses' activities is different between the Italian and the Spanish ICU.
In the Spanish ICU, many activities of daily living (patients' toileting and bathing) are performed by nursing assistants, while in the Italian ICU the nursing staff have to perform all the tasks involved in the everyday direct patient care.
As a result of this different healthcare organization, in the Spanish ICU the nursing staff spend much more time in HAIs surveillance activities. The difference seems to be statistically relevant (Table 4).

SAPS II score
Nurse to Patients ratios

Discussion of Findings
The observers' researchers showed a moderate strength of agreement in the use of both the SAPS II and the NEMS tool.
The nursing workloads and the time spent in NI surveillance during the first shift in both ICUs seem to influenced by the different distribution of health-care activities.
We did not specifically compare the experience of the nursing staff (years of work in ICU), nor the level of education because of the current laws: the Strasbourg Treaty (1967) about the minimal standards for the instruction and education of nurses, and the Treaty of Rome (1957) which provides the right for the free movement of workers within the European Economic Community. Nurses who obtained their title of RN in Europe may work in each State of the European Union.

Implications for healthcare providers
Nursing activities are more effective in reducing the risk of HAIs when performed in accordance with a prevention protocol, so nurses should be allowed to spend more time in NI surveillance activities and to perform evidence-based aseptic work practices.
The presence of nursing assistants in ICUs should be encouraged, because they help reducing the time spent in the everyday direct patient care and the nursing workloads, thus increasing the time spent in HAIs surveillance activities.

Conclusion
The most evident threats to patient safety associated with nursing care occur in hospital inpatient units, especially among critically ill subjects. The strategies of prevention of many NI are well established, however those rules are not currently applied by the heath care workers in their everyday work [14]. Nursing administration and policy most urgently need research conducted with standardized data collected at the patient care unit level. Some authors observed that excessive nursing workload may represent a risk factor for acquiring HAI, followed by severity of clinical condition [15]. In the Spanish ICU, these activities were performed in accordance with the government project "Bacteriemia zero" [13], which is part of a protocol of global strategy of HAI prevention in the ICUs supported by the Joint Commission.
We observed that the presence of nursing assistants, in the Spanish ICU, allows nurses to spent more time in NI surveillance activities and, probably, to be more effective in reducing the risk of infection and colonization through evidence-based aseptic work practices.
This study has important limitations. It involved two cardiac surgery ICUs, consequently, our findings cannot be generalized to all patients undergoing critical care. We did not assess the rate of HAIs occurred during the observation, nor the mean time of stay in hospital due to complications. In addition, our observation was limited to the nursing activities during the first shift of work.