Impact of two bundles on central catheter-related bloodstream infection in critically ill patients 1

ABSTRACT Objective: To evaluate the impact of the implementation of insertion and maintenance bundles on the rates of catheter-related bloodstream infection in an intensive care unit. Method: This is a quasi-experimental, before-and-after study with a non-equivalent control group. During a six-month period, insertion and maintenance bundles for the central venous catheters were implemented. Supervision guidelines were developed to assess compliance with the bundle and catheter characteristics. Results: A total of 444 central catheters corresponding to 390 patients were observed, of which 68.7% were inserted in the unit. The maintenance and insertion bundles reached 62.9% and 94.7% compliance, respectively, and 50.7% of the insertions were supervised. It was possible to observe a 54.5% decrease in the rate of central catheter infection (3.48 vs 1.52 x 1000 days/catheter, p<0.05) when compared with the control group. Conclusion: The simultaneous implementation of insertion and maintenance bundles has a positive impact on the reduction of catheter-related bloodstream infection; therefore it is an efficient alternative to improve the quality and safety of care in high complexity units.


Introduction
Intensive care units (ICUs) are highly qualified to care and treat patients at risk through invasive therapy, procedures and devices such as the central venous catheter (CVC). The CVC is one of the most common devices in the ICU, since it is used to monitor hemodynamics and deliver vasoactive drugs, antibiotics and total parenteral nutrition (1)(2) . Despite its benefits, the CVC can lead to mechanical or infectious complications.
The latter are more frequent and have greater impact on the patient (3)(4) . Catheter-related bloodstream infection (CRBSI) is a complication that can be related to increases in costs, length of stay and morbidity and mortality rates, especially among ICU patients (5)(6)(7) . A recent analysis showed a 2.75-fold increase in hospital mortality and a 2.15-fold increase in CRBSI on ICU patients (8) . Likewise, a study from Argentina found that CRBSI was associated with an additional cost of almost $5,000 and an increase in hospital stay of 12 days for each episode (9) .
There are several risk factors associated with CRBSI, such as: duration of catheterization, number of lumens, femoral access site, excessive manipulation of the CVC, total parenteral nutrition, bacterial colonization at the insertion site, prolonged hospitalization, and others (10)(11)(12) . There are several strategies to prevent CRBSI, and bundles are recognized as one of the most used and most effective for the reduction of CRBSI (10,13) . Bundles can be defined as the systematic implementation of a set of evidence-based practices, usually three to five, that, when performed properly and collectively, can improve patient outcomes (13) . Research on CRBSI prevention demonstrated the effectiveness of bundles, which reduce the incidence of CRBSI by up to 80% (5)(6)14) , reaching a rate of 0 in some cases (4,15) .
The bundles for the prevention of CRBSI include good hand hygiene, preparation of the skin with chlorhexidine gluconate, preference for the subclavian vein, maximal sterile barriers, and daily assessment of the need for the CVC (16) . Thus, basic infection control practices can significantly reduce the incidence of CRBSI, reducing the rate of 6.5 to 46 cases per 1000 CVC days (17) .
In order to promote quality and safety in the care process, health institutions had to explore and adopt practices to minimize risks to patients. An unusual increase in the number of CRBSI cases was observed in the unit under study on the first semester of 2015.
This motivated the creation of a quality assurance plan to reduce and prevent these events, including the implementation of a CRBSI bundle. The objective of this study was to evaluate the impact of the implementation of CVC insertion and maintenance bundles on CRBSI rates in a medical-surgical ICU (MSICU).   (4,16) .

Method
In this study, the simultaneous implementation of two bundles may have compensated for the lack of reports on the insertion bundle and the low compliance with the maintenance bundle. The level of compliance with the maintenance bundle was low compared to other studies (18)(19) . A 95% adherence is suggested in order to achieve a substantial reduction in CRBSI rates, which is usually not achieved (20) . The maintenance bundle, on the other hand, reached high compliance levels for each individual measure and included the daily evaluation of the CVC insertion site, a measure that is not frequent, despite its presence in the bundles that led to a decrease in the CRBSI rate (18)(19) .
Another factor that may have influenced the reduction of CRBSI is the inclusion of daily bathing with 2% chlorhexidine gluconate, which is normally not included in the bundle measures (6,10) . The use of chlorhexidine instead of normal soap is based on reducing the bacterial colonization on the patient's skin, which can enter the bloodstream via the extraluminal CVC route (21) . The chlorhexidine cleaning is recommended when the basic measures for the prevention of CRBSI did not achieve the results expected; however, its use is justified when the CRBSI rates are above institutional limits (11) .
The low level of compliance with the maintenance bundle, especially during the first half of the intervention the implementation and execution of these practices (1) . However, almost 80% of the ICU team was trained on the bundles and received monthly suggestions and comments regarding their effectiveness.
In addition, the first three months of the control  Unlike the insertion bundle, the effectiveness of the maintenance bundle was based on the assessment of clinical records, with no verification of the accuracy of such data. In addition, the monitoring and supervision periods of the bundles did not allow evaluating the compliance and the impact on the CRBSI on mediumterm. Although the literature reports similar periods of surveillance (16,18,24) , other studies have described periods longer than 6 months (7,(25)(26) .

Conclusion
The in the critical processes that determine the occurrence of CRBSI, therefore they are workers that assure the quality and safety of care for the critically ill patient.