Chlorhexidine and gauze and tape dressings for central venous catheters: a randomized clinical trial

OBJECTIVE: to assess the effectiveness of the chlorhexidine antimicrobial dressing in comparison to the gauze and tape dressing in the use of central venous catheters. METHOD: a randomized clinical trial was conducted in the intensive care and adult semi intensive care units of a university hospital in the south of Brazil. The subjects were patients using short-term central venous catheters, randomly assigned to the intervention (chlorhexidine antimicrobial dressing) or control (gauze and micro porous tape) groups. RESULTS: a total of 85 patients were included: 43 in the intervention group and 42 in the control group. No statistically significant differences were found between dressings in regard to the occurrence of: primary bloodstream infections (p-value = 0.5170); local reactions to the dressing (p-value = 0.3774); and dressing fixation (p-value = 0.2739). CONCLUSION: both technologies are effective in covering central venous catheters in regard to the investigated variables and can be used for this purpose. Registry ECR: RBR-7b5ycz.


Introduction
The Central Venous Catheter (CVC) is a device largely used by patients in critical health conditions. Its use, however, may entail many complications, among which is Primary Bloodstream Infection (PBSI), which is the most common bloodstream infection in patients with a CVC for more than 48 hours and that is not related to another site (1) .
A concern to promote measures that reduce PBSI is justified given the high rates of associated morbidity and mortality. A patient affected by this complication will remain hospitalized for another seven to 21 days, with a consequent increase in hospital costs from U$3,700 to U$29,000 and in a risk of death; attributable mortality for this infection is 18% (2) .
In Brazil, it is estimated that 60% of infections are catheter-related PBSI, especially short-term catheters (3) .

Considering the relevance of this figure, the Brazilian
Health Surveillance Agency (ANVISA) has made efforts to prevent this complication. Among the measures adopted is the goal to reduce PBSI rates by 30% in three years (4) . Various facilities in the international context have reported the adoption of severe measures to reduce PBSI rates to zero. Given the impact of this complication for patients and healthcare services, a continuous search for effective prevention is justified (5) .
Despite the large number of PBSI risk factors, these are minimized with the adoption of interventions upon catheter insertion and maintenance such as: implementation of bundle care (hand hygiene, maximum barrier precautions upon insertion, chlorhexidine skin antisepsis, optimal catheter site, and prompt removal) (6) ; maintaining the output opening, occluded with sterile dressing; among other interventions (1,3) .
In regard to sterile occlusive dressings, there are different technologies on the market, such as gauze and tape, transparent polyurethane film dressing, and chlorhexidine antimicrobial dressing. These vary in terms of durability, ease of application, skin reaction, and ability to prevent infections. Gauze and tape dressings and transparent polyurethane film have been widely investigated recently and meta analysis addressing this topic reports that the latter leads to four times as many infections (7) .  (1,4) .
Therefore, this study's aim was to assess the effectiveness of chlorhexidine antimicrobial dressings in comparison to gauze and tape dressings for the following outcomes: prevention of primary bloodstream infections, local reactions, and fixation of dressings.

Method
The study was approved by the Institutional The inclusion of patients in the study followed the scheme shown in Figure 1. We employed a randomized technique in blocks of six patients with a sequence of random numbers generated by Excel ® (8) .
After randomization, the dressings were changed for the first time and the technology to which the patient was assigned was applied. At this point, we collected the socio-demographic and clinical variables, along with catheter data using an instrument defined in the pilot test. Afterwards, the participants were assessed for the collection of the outcome variables as described below.
The outcome variables included: PBSI, fixation of dressing to skin, and local reaction to dressing. In this study, PBSI was the presence of at least one of the following signs and symptoms: fever (>38°); shivering; urine output<20ml/h; systolic blood pressure ≤ 90mmHg; no apparent infection in another site; and medical prescription of antimicrobial therapy for sepsis (4) .
Note that the patients using antimicrobial agents prior to the catheter placement were included in the study provided they did not present clinical diagnosis of sepsis; the PBSI criteria established in the literature did not indicate the need to exclude these patients (4) .  Good fixation was considered to have occurred when the dressing remained intact in more than 75% of the observations. The analysis of variables, local reaction and fixation was performed always at the same time by a staff member previously trained for data collection and, if there was doubt, a second evaluator would assess the variables, as well.
The chlorhexidine antimicrobial dressing was changed every seven days and gauze and tape every 48 hours (2) , or less whenever the dressing edges detached and compromised the dressing integrity or discharge accumulated in the catheter opening (2) . For the catheter puncture, we used the standardized technique adopted by the unit that is based on the care bundle recommended by international clinical guidelines (1,6) . A qualified staff member exclusively performed the dressings using a standardized technique. Patients were monitored up to the end point, which comprised transference to another hospital facility, puncture of a second catheter, or its Inclusion criteria Being 18 years old or older, hospitalized in the ICU ir ASICU; in first short-term CVC less than 24 hours; written consent; no know sensivity to the dressings.   no statistically significant differences between groups.

Exclusion criteria
Most cases of local reaction were characterized by skin hyperemia and maceration. Note that there was one case of a severe reaction to the dressing in the intervention group and for this reason that participant was excluded from the study (Table 3).

Discussion
Catheter-related infection is a complication that greatly influences the morbidity and mortality of patients using CVCs, a fact that justifies unceasing efforts to promote preventive evidence-based interventions. The technologies addressed in this study are safe options for occluding catheter openings due to the low incidence of PBSI in both groups (13.95% -CHD; 11.90% -gauze).
No significant differences were found between the dressings in regard to the occurrence of PBSI (p-value = 0.5170). The incidence of PBSI, however, can be even lower, as shown in a cohort study conducted in seven ICUs that reported an infection rate 6.4% (9) .
One meta analysis and two clinical trials (11)(12) were found in an extensive search of the literature (10) . These studies assessed the ability of chlorhexidine dressings to reduce colonization rates and PBSI. Nonetheless, all studies used transparent polyurethane dressings in the control group. This is the first Brazilian study addressing this new technology that also employs gauze and tape as the control group. Note that this is a safe choice, since the meta analysis included six clinical trials comparing gauze and tape with transparent polyurethane and revealed no differences between these technologies in regard to the incidence of catheter-related infectious complications (7) .
A meta analysis that included eight randomized clinical trials assessing chlorhexidine-impregnated dressing showed that this technology has a tendency to reduce PBSI, though with no statistical evidence (10) . Two new clinical trials were published in 2009 addressing this topic and both showed significant decrease in PBSI among patients using chlorhexidine dressings (11)(12) . The studies, however, used a sponge impregnated with chlorhexidine, which needs to be covered with transparent polyurethane dressing. In this study, the chlorhexidine dressing used Note that the absence of statistically significant differences in the effectiveness of chlorhexidine and gauze and tape dressings in the prevention of PBSI may have been influenced by the fact that only the first CVC of each patient was included in the analysis, since the more frequently patients are exposed to CVC, the higher the risk of complications. Another important factor is related to the short duration of catheter placement, as periods longer that five days are associated with a higher risk of infection.
The only case of infection confirmed by a laboratory was associated with the Candida krusei microorganism.
Hyperemia (20.93% -CHD; 23.81% -gauze) in the catheter opening was the sign most frequently observed in this study, while purulent discharge was the least observed (4.65% -CHD; 2.38% -gauze), though none presented significant association with the occurrence of PBSI. One study conducted with 37 patients who acquired PBSI in a university hospital found the following to be the two most frequent signs: purulent discharge (27%) and hyperemia in the catheter opening (18.9%) (13) . International clinical guidelines recommend daily inspection and palpation of the catheter site to check for signs of inflammation (1) . Data resulting from palpation, however, such as edema (p-value = 0.2993) and pain (p-value = 0.2695) did not present a significant relationship with the PBSI outcome.
The dressings analyzed in this study were associated with local reactions (CHD -39.53%; gauze -45.24%), however, no statistical significant difference was found between the groups. According to the data found, one randomized clinical trial conducted with 21 ICU inpatients also reported a high rate of gauze and taperelated local reaction (60%) (14) . The clinical signs more frequently observed for gauze and tape dressings were skin maceration (28.57%) and hyperemia (21.43%).
Even though the micro-porous tape is hypoallergenic, the signs observed were concentrated in the region in which skin was in contact with the tape and not in the region of direct contact between the gauze and the patient's skin.
In regard to the chlorhexidine dressing, the part There was one case of a severe reaction to CHD, a fact that caused discontinuing the patient in the study.
We note, however, the low incidence of severe reactions to the dressing, a fact that would impede its widespread use and reaffirms the view that this dressing should be prescribed only for patients who do not present known hypersensitivity to chlorhexidine antisepsis. There is one study corroborating these findings, as it reports that chlorhexidine is well-tolerated by patients (15) . clinical trial assessing gauze and tape dressing shows that 50% of the dressings used in the analyzed catheters presented poor fixation and required early changing (14) .
In this study, the rates of poor fixation found for both the gauze and tape and chlorhexidine dressings were lower than that reported in the literature. Note that the good fixation of dressings found in this study is a factor relevant for the maintenance of occluded dressings, which favors the reduced colonization of peri-catheter skin.
Limitations of this study include the fact that the

Conclusion
The objective was to assess the effectiveness of the chlorhexidine antimicrobial dressing in comparison to the gauze and tape dressing for the outcomes: prevention of primary bloodstream infections, local reactions, and the fixation of dressings.
This study shows that, in a unit that adopts catheter bundle care, the chlorhexidine antimicrobial dressing is not effective in reducing PBSI when compared to the gauze and tape dressing. In regard to the occurrence of local reactions and the fixation of dressings, no statistically significant differences were found between groups. A high rate of local reaction was observed in both groups, as well as the occurrence of a severe skin reaction in the case of one patient using the chlorhexidine dressing.
Both the technologies addressed in this study are effective in covering CVCs in regard to the assessed outcomes and can be employed for this purpose. We recommend further research to consolidate scientific evidence for the use of chlorhexidine dressing in the prevention of primary bloodstream infections in critical patients.