Received Date: 01/12/2020; Accepted Date: 17/12/2020; Published Date: 24/12/2020
Visit for more related articles at Research & Reviews: Journal of Nursing and Health Sciences.
Introduction: The central venous catheter use can put risk to patients’ safety in the intensive care unit. Nursing care requires continuous assessment to ensure adverse events reduction.
Objective: To demonstrate the evidence that nursing care in the intensive care unit reduces adverse events related to patients’ safety during insertion, maintenance, and removal of a central venous catheter.
Material and Methods: Systematic review and meta-analysis. Eight databases were searched using key terms. Two researchers independently assessed data. The Kappa coefficient was calculated. The risk of bias and the quality of the evidence were analyzed.
Results: Eight studies were included. The Kappa coefficient was 0.64 and 0.73. Although the quality of the evidence was very low, nursing care for patients’ safety in the maintenance of central venous catheters were identified. There was no significant effect of the nursing care bundles related to central venous catheter maintenance on central line-associated bloodstream infection. The methodological quality of the included studies was moderate to severe in observational studies and a high risk of bias for the randomized design.
Conclusion: Nursing care, when appropriate and clinically evaluated, contributes to critical patients' safety reducing adverse events at all the moments during the use of central venous catheters.
Review, Nursing care, Nurse, Catheters, Intensive Care Units.
Nursing care can determine different outcomes for patients in intensive care units (ICUs). The technological profile of these units allied to the clinical patient’s condition contributes to a dynamic and complex system, conducive to the occurrence of unfavorable outcomes . Non-tunneled devices, like central venous catheters (CVCs), is a technology widely used in critical patients’ treatment; however, the CVC use is also associated with adverse events .
Adverse events are episodes of incidents that result in damage to the patients . Bloodstream infections (BSI) are the main problems related to CVC, but extravasation/infiltration, hematoma/ecchymosis, catheter embolism, gas embolism, deep venous thrombosis, catheter obstruction, and catheter rupture can also occur . Therefore, the CVC use is a risk to patients’ safety in ICU, because of that, continuous assessment of the care provided is necessary.
The health care for BSI prevention and control are compiled in conjunction with care activities directed to the patient care team, as in the case of care bundles. They are a set of interventions developed based on guidelines and applied in the clinic for a defined population of patients and environments . The care bundles were initially introduced by the Institute for Healthcare Improvement (IHI) in 2001 with a focus on improving care in the ICUs . They aimed at the improvement of critical care processes towards high levels of reliability, with a consequent improvement in patient outcomes .
In 2011, was published guidelines recommended care bundles aiming at reducing intravascular catheter-related infections. These recommendations drove changes in the clinical practices of the entire patient care team, especially for healthcare personnel who insert intravascular catheters and for persons responsible for surveillance and infection control .
The management of adverse events is important in health services since a patient with damage resulting from an adverse event has a higher economic cost, a longer hospital stay, a higher probability of readmission in 30 days, and a higher probability of death .
Studies have shown that nursing care contributes to better patient health outcomes, therefore being a care quality predictor [10,11]. Thus, it is important not only to identify the care that is provided by nurses for patients with CVC but also to demonstrate the evidence about these nursing care for patient safety in ICU, being, therefore, the aim of this systematic review.
This review was conducted using the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) for structuring the protocol . The protocol was previously registered on the International Prospective Register of Systematic Reviews platform (PROSPERO - number CRD42020152630) .
The acronym PICO was used to construct the guiding question for this review. In this, the letter P represents the population/ participant; the letter I represents the intervention/procedure; the letter C represents the comparison (optional in the studies); and the letter O represents the outcomes . This review was guided by the following question: “What is the evidence that the nursing care in the intensive care unit reduces adverse events related to patient safety with respect to the insertion, maintenance and removal of central venous catheters?”
Databases were systematically searched on the 12th of November, 2019. Databases included MEDLINE accessed by the National Library of Medicine (PubMed) and by the Virtual Health Library (VHL); Latin American Health Sciences Literature (LILACS); Nursing Database (Base de dados em Enfermagem, in Portuguese - BDENF); Cumulative Index to Nursing and Allied Health Literature (CINAHL); Scopus; Web of Science (WOS); Cochrane Library; and Scientific Electronic Library Online (SciELO). Details of the PubMed search strategy are presented in Table 1.
|PubMed||("Nursing Care" OR nurses AND catheters OR "Catheterization, Central Venous" OR "Central Venous Catheters" OR "Central Venous Catheter" OR "Vascular Access Devices" OR "Vascular Access Device" OR "Vascular Access Ports" OR "Vascular Access Port" OR catheterization OR catheterizations OR cannulation OR cannulations AND "Intensive Care Units" OR "Intensive Care Unit" AND adult OR "Young Adult" AND ("2010/01/01" [PDAT] : "2019/11/11" [PDAT]) AND (English [lang] OR Portuguese [lang] OR Spanish [lang]))|
Table 1. Search strategy applied in the PubMed database.
The following inclusion criteria were adopted: Randomized Clinical Trial (RCT) study; quasi-experimental study; analytical observational studies; in English, Portuguese or Spanish; published from 2010 to 2019 (date prior to collection); studies with reduction of adverse events or patient safety as outcome; studies that present nursing care to adult patients with CVCs in ICUs, or with indication of installation and/or removal of the catheter.
The exclusion criteria were: theses, dissertations, editorials, review studies; studies not fund in full length; studies with pediatric or neonatal populations; studies with patients using: arterial catheters, or catheters for hemodialysis, or catheters for plasmapheresis; studies with patients using catheters inserted in peripheral veins; and use of a tunneled catheter.
The reference focus defined for the search period from 2010 to 2019 was given by the publication of the guidelines in 2011 on the adoption of care bundles .
References were exported to the EndNote Basic™ (Clarivate Analytics). Those indexed in more than one database were excluded. Two researchers independently examined the titles and/or abstracts. For this step, an electronic tool was used to enable independent reading and evaluation by researchers - Rayyan® .
The selections agreement level was analyzed using Kappa coefficient with a 95% Confidence Interval (CI). The Kappa coefficient allows to exclude the hypothesis of concordant selections due to chance . The following classifications were adopted for the numerical indices identified: < 0.00, poor agreement; 0.00 - 0.20, slight agreement; 0.21 - 0.40, reasonable agreement; 0.41 - 0.60, moderate agreement; 0.61 - 0.80, substantial agreement; and 0.81 - 1.00, almost perfect agreement .
Full-texts of potential articles were reassessed after the first screening. The pair of researchers independently read full-texts of these articles to refine the final selection. The results obtained in this stage were also submitted to the agreement level analysis using Kappa coefficient, with 95% CI. Disagreements were resolved based on the evaluation by a third researcher.
Data were extracted in a spreadsheet built in Microsoft Excel®, version 2019, with the following headings: first author; year of publication; country; title of the study; study aim; study design; description of the nursing care; moment of the nursing care; and outcomes of the nursing care.
The synthesis of evidence was generated from the results of all selected studies. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) was applied . The grading of the quality of the evidence determined by the evaluators can reach one of four different levels by applying the GRADE, namely: high, moderate, low and very low . The online tool GRADEpro GDT was used for this evaluation and summarization .
Applying the GRADE, five criteria allow reducing the evidence level: risk of bias, inconsistency, indirect evidence, imprecision, and publication bias. By the other hand, three criteria allow increase the degree of confidence in the effect estimate, and thus, raise the level of quality of the evidence in studies that were not previously downgraded by the above criteria. These criteria are great magnitude of effect, dose-response gradient, and factors of residual confusion .
We used the Risk of Bias in Non-randomized Studies - of Interventions (ROBINS-I) tool for the observational studies . The ROBINS-I tool makes it possible to evaluate seven domains: bias due to confounding; bias in selection of participants into the study; bias in classification of interventions; bias due to deviations from intended interventions; bias due to missing data; bias in measurement of outcomes; and bias in selection of the reported result. In this case, risks of bias that can occur from the preintervention, intervention and post-intervention stage of the study are evaluated .
The option to use ROBINS-I is available in the GRADEpro GDT; its application influences the final estimate of the evidence, giving greater weight to observational studies, with the following possibilities for grading the risk of bias: low, moderate, serious, critical, and no information .
We used The Cochrane Collaboration - Risk of Bias (RoB2) tool to assess the risk of bias in RCTs. This tool allows the evaluation of the randomization process, deviation from the intended interventions, the missing outcome data, measurement of the outcomes, and selection of the reported results. In this tool, the possibilities for classifying the risk of bias are as low, high or uncertain risk of bias .
For studies considered comparable and relatively homogeneous in terms of design, interventions, and outcomes, a metaanalysis was performed using Review Manager 5.4®, summarizing the results using the Mantel-Haenszel model, Chi-Square test and interclass correlation statistics (I2) . Significant heterogeneity was defined as I2 > 50% or p < 0.05 with the Mantel-Haenszel Chi-Square test. The results were presented with 95% confidence intervals (95% CI). Statistical significance was considered p < 0.05.
The report of this review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines . In addition, tables and boxes were adopted to present the results as to the characteristics of the studies, as well as the evaluations obtained from the GRADEpro GDT. The classifications from the application of the tools for assessing the risk of bias in the studies were presented in the form of graphs from an electronic tool - robvis .
Databases searches identified 613 records. Were excluded 224 of these due to duplication and 331 were excluded in the first screening. A Kappa coefficient was 0.64 of the agreement. Fifty-eight articles were read in full length, of which nine were included and 49 were excluded, as shown in detail in Figure 1. Kappa coefficient of 0.73 agreements.
Figure 1: Flowchart of the route to identify the studies and selection for inclusion. Note: LILACS: Latin American Health Sciences Literature; BDENF: Nursing Database (Base de Dados em Enfermagem, in Portuguese); CINAHL: Cumulative Index to Nursing and Allied Health Literature; Cochrane: Cochrane Library; SciELO: Scientific Electronic Library Online (SciELO).
A study, with RCT design, provided results for the publication of two articles in different journals. Therefore, although nine articles are counted, they correspond to eight studies, which were used for the synthesis of evidence.
The characteristics of the eight studies are summarized in Table 2. The eight studies were developed in four different countries, of which one was in Brazil (1/12.5%). The majority was developed in the United States of America (4/50.0%), followed by Turkey (2/25.0%), and India (1/12.5%). With the exception of one RCT (12.5%), all other studies were observational, of before and after type (7/87.5%). The works were published from 2012 to 2018, the majority in 2012 and 2014, with two publications (25.0%) in each year. In 2018, 2017, 2016 and 2015, there was one (12.5%) publication each year.
|Author/ Year/ Country||Study aim and design||Design||Nursing care||Care Moment||Outcomes|
|Fox et al., 2015/USA ||To investigate a new protocol of patients’ HH to reduce infection rates and improve the compliance with HH among ICU nurses.||Observational||To apply the patient’s HH protocol with 2% chlorhexidine towels, 3 times/day||CVC maintenance||*1.1 to 0.50 CLABSI (p = 0.64) 9 months without infection|
|To perform HH before and after contact with the patient||CVC maintenance|
|Kahn et al., 2014/ USA ||To determine whether remote screening by nurses and the request for EBP, using an electronic health record, could affect the provision and results of care in the ICU.||Observational||To perform remote monitoring, from the EHR, of the need to assess the continuity of use of the CVC||CVC maintenance||*0.72 to 0.77 CLABSI. Infection number from 16 to 32 (p = 0.84)|
|Kaya et al., 2016/ Turkey ||To determine the effect of a Nursing Care Protocol on the prevention of CLABSI in a neurosurgery ICU.||Observational||To apply a nursing care protocol||CVC maintenance||CLABSI number from 9 (11.3%) to 4 (5.0%) (p = 0.14)|
|Munoz-Price et al., 2012/USA ||To determine the effectiveness of a series of interventions for reducing CLABSI rates.||Observational||Phase 2: Friction of intravenous ports for 15 seconds with single-use chlorhexidine swabs||CVC maintenance||Each ICU (A, B and C) adhered to a number of different interventions; ICU A: Statistically significant CLABSI reduction in all phases; ICU B: No significant CLABSI reduction between phase 1 (pre-intervention) to phase 2; ICU 3: Significant CLABSI reduction from phase 1 (pre-intervention) to phase 3.|
|Phase 3: Daily baths of patients with 2% chlorhexidine wipes||CVC maintenance|
|Phase 4: Daily rounds of the Nursing Manager and the Bedside Nurse on each patient to ensure compliance with a checklist of goals||CVC maintenance|
|Pedrolo et al., 2014/ Brazil [28,29]||To evaluate the effectiveness of the chlorhexidine antimicrobial dressing, comparing it with gauze and tape; to identify factors related to the occurrence of infection, local reaction and poor fixation of CVC dressings.||RCT||Transparent adhesive dressing integrated into a gel pad containing 2% chlorhexidine gluconate; Daily assessment; Change every seven days or as needed.||CVC maintenance||CLABSI number from 5 (11.9%) to 6 (13.9%) (p = 0.51). There was good fixation and high incidence local reaction to dressings (variables without statistically significant difference). Significant association of CVC permanence time > 5 days and infusion of blood components and PBI|
|Gauze dressing and microporous tape; Daily assessment; Change every 48 hours or as needed.||CVC maintenance|
|Pfaf et al., 2012/ USA ||To compare the efficacy of a new one-piece occlusive dressing with chlorhexidine gluconate, with a dressing plus a chlorhexidine gluconate patch, in maintaining low rates of CLABSI in the ICU.||Observational||To change one-piece occlusive dressing with chlorhexidine gluconate||CVC maintenance||* 0.53 to 0.52 CLABSI rate|
|To change two-piece dressing with chlorhexidine gluconate||CVC maintenance|
|To apply a care bundle||CVC maintenance|
|Sahni et al., 2017/ India ||To establish the incidence of VAP and CLABSI, intervening in the education, training and feedback of nurses and analyzing its impact on the incidence of VAP and CLABSI.||Observational||To perform HH||CVC maintenance||*7.90 to 1.73 CLABSI rate (p < 0.05)|
|To apply a care bundle||CVC maintenance|
|Yazici et al., 2018/Turkey ||To evaluate the effectiveness of a care bundle with the objective of preventing the three most frequent infections in the ICU.||Observational||To apply a care bundle||CVC maintenance||* 8.90 to 9.90 CLABSI rate (p = 0.24). CLABSI number from 10 to 10.|
Table 2. Summary of the included studies (n = 8).
All studies CLABSI as the primary outcome. However, other outcomes were evaluated, such as the ability of the dressing to develop a local reaction (characterized by maceration, hyperemia, flaking or itching in the region of contact between the dressing and the skin) and fixation of the dressing to the skin. In general, the nursing cares goals were to reduce the CLABSI, although some aimed at zero rates. The units of measurement of infection were presented both in percentages and also by calculating the rate per 1,000 days of catheter use. All nursing cares were directed to CVC maintenance.
Among the identified nursing cares, care bundles were more frequent. In three studies, bundles were implemented by nurses to patients using CVC [30-32]. However, only one study described the nursing cares, namely: daily assessment of the need for continuity of the CVC use; daily assessment of the need for CVC dressing; evaluation of the dressing as to proper application; identification with the date on the dressing; evaluation of the need to replace fluid sets .
We conducted a meta-analysis to assess the nursing care bundle. We compared with before the intervention. Two studies provided the data [26,32]. The other studies were not included due to heterogeneity, in addition to the absence of data that could be compared. The results pointed in favour of the intervention. In other words, there was a reduction due to the implementation of nursing care bundles. However, this finding was not statistically different between the groups (RR 0.74; 95% CI 0.38-1.43; p = 0.37) as presented in Figure 2.
Care protocols were also identified in the studies analyzed. They included: the Hand Hygiene Protocol of patients with 2% chlorhexidine towels, three times a day ; and Care protocol for the CLABSI prevention in a neurosurgical ICU built-in line with the suggestions of the CDC .
Two different monitoring possibilities were obtained from the studies: remote monitoring and on-site monitoring. It is also noteworthy that remote monitoring was performed by nurses with experience in the ICU (minimum three years), by screening from the electronic medical records , while on-site monitoring was conducted by a pair of professionals, the Nursing Manager and the Bedside Nurse .
In addition to the on-site monitoring by nurses (manager and clinician), in the same study, other care measures for maintaining the catheter were identified. The following precautions were taken: friction of intravenous ports for 15 seconds with single-use chlorhexidine swabs; and daily baths with 2% chlorhexidine wipes .
Different dressing coverages at the insertion site of the CVC were compared, highlighting chlorhexidine gluconateimpregnated dressing. All dressing evaluated were effective for the related outcomes. The two types of dressing impregnated with chlorhexidine gluconate (a 1-piece group and another 2-piece group) maintained low rates of BSI, although the 1-piece coverage was economically more viable . In addition, both the dressing integrated with a gel pad containing 2% chlorhexidine gluconate and the gauze dressing and microporous tape contributed to the CLABSI reduction, local reaction, and resulted in good dressing fixation [28,29].
In the analyzed literature, other factors were significant association with the occurrence of local reactions: the catheter’s length of stay greater than five days; the number of dressing changes; and the insertion site. In addition, there was a significant relationship between poor fixation and more than two changes with chlorhexidine dressing, and between catheters inserted in the jugular vein and dressing in gauze .
Nursing care regarding the management of the use of catheters was identified. There was a significant association with the administration of blood components and the occurrence of BSI, with a 10.29-fold higher risk in the group that used a chlorhexidine antimicrobial dressing; in addition, catheter’s stay of more than five days had an eight-fold higher risk of developing an infection in the group that used gauze dressings and microporous tape .
Only three studies provided data regarding nurses’ adherence to care measures. As for hand hygiene before getting in contact with the patients, there was an increase in adherence of 66% after the intervention, compared to 35% before the intervention; after contact with the patients, the increased from 66% to 79% . As for the friction of the intravenous entry ports with chlorhexidine swabs, there was 100% adherence; however, the median time of friction was nine seconds . Worsening of adherence related to care with dressing with CVC insertion was identified. The dates of exchanges were not recorded in 51.7% of the observations during the implementation of the care bundles. Furthermore, the nurses’ adherence was monitored for three months after implantation of the care bundles, a decreased to 30.8%, in the period of implantation of the care bundles, adherence was 47.7% .
The risk of bias was assessed for each study and is presented in Figure 3. In general, weaknesses in reports were identified among the studies. The RCT was classified as having a low risk of bias in the first three domains evaluated, related to randomization, deviation from the intended interventions, and missing data; and high risk of bias was detected for the domains of measuring the outcome and selective reporting of results, resulting in a high risk of bias. As for the observational studies, three presented serious limitations related to bias due to confounding, and this was decisive for the final estimate, despite better results in the other domains. Four studies had a general classification as a moderate risk of bias. Limitations were identified in bias due to confounding in three studies, bias in the selection of participants in one study; the bias in the measurement of outcomes in two studies; and bias in the selection of the reported result in one study.
Although the studies answer the same question given from the PICO strategy, they were divided into two groups due to the different methods and designs for evaluation by the GRADE as presented in Table 3. Substantial limitations in observational studies were due to inconsistency - heterogeneous studies; and indirect evidence because different interventions were observed. In the RCT evaluation, limitations were identified in the risk of bias according to RoB 2; and inaccuracy due to the small sample size.
|Number of Studies||Risk of bias||Inconsistency||Indirect evidence||Inaccuracy||Others*||Level of evidence|
|7||Serious||Very serious||Very serious||Serious||None||⨁◯◯◯
|1||Very serious||Not serious||Not serious||Very serious||None||⨁◯◯◯
Table 3. Evaluation of the GRADE criteria and level of evidence, using the GRADE pro GDT tool.
To our knowledge, this is the first systematic review to demonstrate the evidence that the nursing care in the ICU reduces the occurrence of adverse events related to the safety of patients using CVC, related to its insertion, maintenance, and removal. In this study, it was found that nursing care contributes to the safety of patients using CVC in ICUs, but the evidence is supported in very low quality.
The assessment of the risk of bias with the use of validated tools revealed weaknesses in the methods adopted in the studies. Most studies applied a before and after design, which is prone to bias, while only one RCT was included in the analyses. Randomized clinical trials are more rigorous in terms of design when compared to other study types . Therefore, these results limit the possibility of generalizing the findings due to the internal factors of the included studies.
Given the possibility of comparing the studies, a meta-analysis was carried out to investigate the effect of the application of nursing care bundles in the maintenance of catheters, in which, there was a trend towards a non-significant reduction in CLABSI. Considering they are sets of care that can be applied in their entirety or not, depending on the demand for care, conducting a subgroup analysis for each component is important. Comparisons between the adherences of nurses to each item of care bundles are required.
Studies are developed to assess the impact of implementing care bundles within the multidisciplinary team. The following practices are present in their composition: hand hygiene before insertion of the catheter; use of a sterile barrier with maximum precautions, such as a cap, mask, apron, gloves, and sterile field to cover the entire body of the patient; and use of chlorhexidine for skin preparations (with spontaneous drying before insertion of the catheter); avoiding the femoral vein, if possible, and removing catheters when they are no longer needed; and the inclusion of kits composed of materials necessary for the insertion of central catheters .
In the case of nursing care, these care bundles encompass techniques of the profession. In this study, the following items were identified in the care bundles: daily assessment of the need for continuity of use of the CVC; daily assessment of the need for a CVC dressing; evaluation of the dressing as to proper application; identification with the date on the dressing; evaluation of the need to replace fluid sets . Other activities are also listed in the literature, related to these bundles: education and training in relation to CVC management; performing hand hygiene; use of sterile gloves (or no-touch technique); dressing with 0.5% alcoholic chlorhexidine or isopropyl alcohol, and spontaneous drying; use of the pulsatile discharge technique when the blood returns to CVC; and CVC lumen exclusive for the administration of parenteral nutrition .
All the nursing care identified in the studies were directed to maintaining the catheter. This finding is in line with the statement that most complications related to the catheter occur during the period of maintenance of the device .
Daily assessment of the need for continuity of use of the CVC is part of the best practices for the maintenance of the catheter. In the results observed in this study, it was identified that monitoring - remote or bedside - did not constitute an isolated intervention; resources such as the clinical experience of the nurse or discussion by a pair of professionals were associated. Monitoring, also called daily rounds, are surveillance to monitor the achievement of stipulated goals . When performed remotely is called telemedicine in the ICU (Tele-ICU) . It is a practice used not only by nurses but also by other team professionals . It is shown that interventions to reduce the unnecessary use of CVCs contribute significantly to reducing the BSI rates in adults [35,36].
As for adverse events, CLABSI was the primary outcome assessed in all studies. Although several adverse events are reported in the context of patients using CVC, the impact of Healthcare-Associated Infections (HAIs) has resulted in different health services and directed the focus of publications on this theme . The scenario of this event related to CVC is also observed in Brazil and has caused changes in the clinical practice of nurses. Among the main changes identified in the consulted literature, there is the implementation of evidence-based quality improvement projects .
The improve care practices for patients using CVC in the ICU are especially for the prevention of infection. In general, as identified in the articles included in this review, the goals of the care institutions are directed at achieving zero infection rates, based on a gradual reduction of events . The maintenance of these rates at zero level does not yet occur in many health institutions, mainly in ICUs, there is a need to manage other factors such as the behaviors of adherence of professionals inserted in ICUs.
In the analyzed literature, we observed an assessment of rates of adherence of nurses, which, in general, were high, contributing to the surveillance of these infections. Studies have low rates of CLABSI when the adherence of professionals to the guidelines is greater than or equal to 95% . Therefore, it requires that the number of patients is planned known . In Brazil, adherence rates are considered high . It is also registered in the literature that, there is an inconsistency between the knowledge of the nursing team and adherence to practices; there is a need to implement continuing education for the control and prevention of these infections [40,41].
Only knowing the recommendations for clinical practice does not imply adherence by professionals. As observed in the literature included in this review, there was high adherence to the friction of catheter connectors; in turn, the recommended maximum friction time was not practiced . This results in compromised adherence when time estimates are needed for activities.
Other factors directly influence patient safety and should be considered when dealing with CLABSI. Studies indicate that the lack of nurses and the lack of an appropriate organizational culture significantly interfere in better results as to the incidence of CLABSI . It is not enough that nurses have high adherence to the guidelines and are committed to quality care, it is also important that health service managers ensure structural and organizational conditions in health services to offer safe care to patients.
Although in compliance with the recommendations, health services organize strategies to improve clinical practice based on their realities. Adaptation of protocols was observed, such as the Hand Hygiene Protocol of patients with 2% chlorhexidine towels . In general, however, studies evaluated not only hand hygiene but also a daily bath with chlorhexidine. Regarding HAIs, a meta analysis allowed to suggest that the daily bath of ICU patients with chlorhexidine reduces the risks . With regard to CLABSI, the evidence also supports this indication .
Regarding the maintenance of catheters related to dressing changes, in the literature consulted, the studies were directed to evaluations of nursing care technologies, especially the dressing of the insertion site [28-30]. Care related to changing dressings of the insertion site of catheters in Brazil is the exclusive responsibility of nurses in the ICU. Including patient guidance on the procedure, prediction of materials with systematic analysis of the necessary and available resources, and assessment of the individual needs of the patients, which are steps contemplated within the planning of the nursing care. Nursing care can be divided into activities relevant to the preparation of the patient and to the preparation of the procedure; dressing and disposal of materials; and registration of the procedure .
No studies were identified with nurses' care for the patient, directed at the time of insertion or catheter removal. With regard to the CVC removal, nurses work to prevent complications that may occur in the act of catheter removal or later, after the procedure. The occurrence of adverse events during the scheduled removal of the CVC is uncommon [46,47]. However, when they occur, these outcomes have a high degree of clinical importance, being in some cases fatal [46,47].
The nursing care include: to placed patients in the supine position, or in the Trendelenburg position to prevent air embolism [46,48]; to educate the waltz maneuver in situations where the patient is able to cooperate; or to remove the catheter during the patient’s active expiration [46,48]; to interrupt the CVC removal due to the occurrence of physical resistance and request evaluation by an intensive care physician ; to apply digital compression to the ostium immediately for five minutes to ensure hemostasis, and to prevent the entry of blisters [46,48]; to ensure the patient supine position for 30 minutes after CVC removal [46,48]; after stabilization, apply a sterile fully occlusive adhesive cover for 24 hours or longer (72 hours) with regard to ensuring hemostasis . Although there is no consensus about the duration of the occlusive dressing, procedures should be planned with a view to minimizing the risk of air embolism, pneumothorax, and secondary hemorrhage .
Critical patients require that the conduct of health services is safe, as well as the care provided by professionals, based on high-quality evidence. The overall quality assessment of the methods applied in the studies indicated a very low level of quality of the evidence, according to the GRADE applied to the set of studies. For this level of quality of evidence, less confidence is attributed to the estimated effect generated, given the possibility of difference in the effect when results are generated from studies with rigorous methods .
Research with a high level of evidence is required in the context of health care, especially to support therapies for critically ill patients, given the vulnerabilities and complexity of the level of health care. Randomization, initially, corroborates a high level of quality of evidence for RCTs [18,49]. It occurs in the methods of these studies and ensures protection against confounding of prognostic factors between the compared groups (control and intervention) . Randomized clinical trials are therefore used to insert evidence-based practices, being this study design recommended to be used by nurses in clinical research to assess the effects of nursing care on patients.
In this study, evidence was found that nursing care in the ICU reduces adverse events related to patient safety when using CVC. However, in the results obtained from the analyses applying GRADE, the evidence was classified as having very low quality. These findings are mainly due to the study design and the content described in the method of the analyzed articles. Also, the fact that the evidence was very low quality may be associated with the low number of randomized clinical studies. In addition, the nursing care presented in the analyzed articles was restricted to maintaining the CVC with regard to reducing BSI, local reaction, and dressing fixation. The results of this review are contributions to the clinical practice of nurses and managers of highly complex services. In the case of institutions that carry out partnerships for the development of research, it is recommended that the design is of the randomized clinical trial type, and with planning and development with a low level of bias and a high level of evidence.
This is supported by Coordination for the Improvement of Higher Education Personnel (CAPES), Brazil, Finance Code, 001.