Printer Friendly

Hemodialysis catheter outcomes pilot study: no dressing coverage with prescribed showering.


To review the outcomes of a hemodialysis catheter pilot study on dressing coverage with prescribed showering.


1. List three interventions noted to decrease catheter-related bloodstream infection rates.

2. Compare the research outcomes for the hemodialysis catheter groups: the shower group versus the non-shower group.

3. Discuss the implications of this research study for nephrology nursing.

Care of hemodialysis (HD) catheters is an important aspect of nephrology nurses' responsibility when providing care for individuals undergoing incenter HD. One aspect of this responsibility is educating patients on catheter care and safe catheter management. However, there is a lack of evidence regarding the best catheter care practices. To date, research has primarily focused on cleaning products for exit sites, HD catheter types, antibiotics, catheter dressings, and needleless connectors (Allon, 2004; Bakke, 2010; Betjes, 2011; Dinwiddie & Bhola, 2010; Lok et al., 2003; O'Grady et al., 2011). Hence, nurses have been educating patients and caring for HD catheters based on tradition and little scientific evidence. In addition, many patients have developed their own personal HD catheter care techniques, which they may not share with the staff. Anecdotal evidence indicates that despite instructions not to shower, many patients are showering, and may or may not change the wet dressing (Kear, Evans, Hain, Schrauf, & Dork, 2013). This increases the risk for catheter-related bloodstream infections (CRBSIs), which may lead to less than optimal outcomes.

There is a need to develop evidence-based protocols to support optimal patient outcomes, such as reduced HD catheter exit site. tunnel, and CRBSIs. In preparation for this study, the primary investigator (PI) contacted the Centers for Disease Control and Prevention (CDC) to solicit an opinion regarding catheter care. Dr. Alexander Kallen recommended evaluating the effect of no dressing coverage for well-healed tunneled HD catheters (A. Kallen, CDC, personal communication, July 2011) because this was highlighted as a gap in the recent CDC guideline for the prevention of intravascular catheter-related infections (O'Grady et al., 2011). The CDC guideline stated that the necessity of an exit site dressing on well-healed cuffed tunneled central venous catheters (CVCs) was an unresolved issue (O'Grady et al., 2011).

The no-dressing coverage was considered an interesting topic to the investigators, but they also wanted to evaluate the safety of showering because most patients with an HD catheter request to shower. Further, there was no evidence to base an answer on the safety of showering. The CDC 2011 guidelines for the prevention of intravascular catheter-related infections recommended not submerging CVCs in water, but stated that showering could be allowed if the CVC and the CVC connecting devices (hub areas) were protected with an impermeable cover while showering to reduce the likelihood of introducing organisms into the catheter (O'Grady et al., 2011).


The purpose of this study was to explore the effect of a prescribed showering protocol on CRBSI rates, tunnel infection rates, and exit site infection rates in individuals undergoing incenter hemodialysis and using an HD catheter; and to evaluate the effect of not using a catheter exit site dressing on CRBSIs, tunneled catheter infections, and exit site infections. See Table 1 for definitions.

Research Questions

This study asked two questions:

* In participants undergoing incenter hemodialysis (HD) with a chest-tunneled HD catheter that is well healed (over one month since placement), does the absence of a dressing over the catheter insertion site impact the rates of CRBSIs, tunnel infections, or exit site infections as compared to nationally published CRBSI, tunnel infection, or exit site infection rates over a six-month study time?

* In participants undergoing incenter HD with a chest-tunneled HD catheter that is well healed (over one month since placement) with all participants using protocol specific hub care, exit site care, and a no-exit site HD catheter dressing option, does prescribed showering impact the rate of CRBSIs, tunnel infections, or exit site infections as compared to HD participants who have the same HD catheter care over a six-month study time but do not shower?

Literature Review

No Dressing Coverage Literature

A literature search yielded only two published studies utilizing a no-dressing intervention for CVC exit site care (Olson et al, 2004; Petrosino, Becker, & Christian, 1988). Petrosino et al. (1988) studied exit site care infection rates among adult oncology patients comparing gauze, two different transparent dressings, and a no-dressing protocol. This study showed no statistical significance in infection rates between the various dressing groups. Olson et al. (2004) suggested that dressings might be a source of infection and recommended development of a no-dressing protocol. In their opinion, the cleansing technique for catheter care is the most important variable in preventing infection.

Olson et al. (2004) hypothesized that eliminating the CVC exit site dressing might reduce CRBSIs. Seventy-eight (78) adult oncology patients (including patients receiving bone marrow transplants) with healed (defined in this article as catheter in place over three weeks) CVC catheters were followed until they developed a febrile episode. The researchers were unable to show differences between the catheter-related sepsis episodes in the dressing and no-dressing group (Olson et al., 2004). This study demonstrated that not using dressings did not increase the infection rate.

Protected Showering Literature

Dunbar Ivy et al. (2009) analyzed CRBSI rates in pediatric patients with CVCs that were used for the administration of intravenous (IV) medications while allowing the patients to shower. Dunbar Ivy and colleagues (2009) noted that the catheter hubs did not provide a water-tight seal, which could allow the shower water to seep into the connections, possibly increasing the risk for waterborne gram-negative bacteremias. They incorporated a closed-hub system using a commercial device for hub protection and also incorporated a sealable plastic wrap to waterproof the hub connectors for showering. A small piece of this sealable plastic was wrapped and pressed around the catheter hubs, resulting in a more water-tight seal before showering. There was no description of their shower technique or exit site shower protection. The article noted that the hubs were kept dry using the Press 'n Seal[R]. This cost-effective product costs less than $4.00 and lasts well over six months. The pre-hub protection intervention bacteremia infection rate in the Dunbar Ivy et al. (2009) study was 1.04 episodes per 1000 catheter days and 0.24 infections per 1000 catheter days after the protected hub intervention.

Several other studies have also noted that catheter hubs are the potential source of entry for pathogens (Beathard & Urbanes, 2008; Frasca, Dahyot-Fizelier, & Mimoz, 2010; Kallen, Arduino, & Patel, 2010). The Kidney Disease Outcomes Quality Indicators (KDOQI) vascular access guidelines note that "good practice and attention to hub care" can significantly reduce catheter related bacteremias by 4-fold (National Kidney Foundation [NKF], 2006, p. S207).

Interventions to Reduce CRBSI Literature

Lok et al. (2003) published an exit site infection rate of 1.02 per 1000 catheter days and CRBSI rate of 0.63 per 1000 catheter days after using polysporin topical ointment at the exit site of HD-tunneled catheters. Bakke (2010), in a prospective study with 187 patients and 4752 catheter days, achieved a catheter-related infection (CRI) rate of 0.2 infections per 1000 catheter days by incorporating the 2002 CDC guidelines (O'Grady et al., 2002). In patients cared for in the year prior to implementing the CDC guidelines, the CRI rate had been 1.7 infections per 1000 days. The new protocol utilized the 2002 CDC guidelines for catheter insertion technique, hand hygiene, chlorhexidine in alcohol for hub and exit site care, and ongoing staff education. The HD catheter cover dressing used in this study was a 2x2 gauze and tape. The Bakke (2010) study also did a cost comparison of usual care to best practice care, comparing the cost of CRIs in the year pre-guideline care to the year using the guidelines for care and estimated that the cost savings for preventing nine CRIs in their dialysis unit during the post-guideline care period was $141,606. It was unclear in the Bakke (2010) study if all infections (exit, tunnel, and CRBSI) were grouped into the CRI rate.

The current study's infection rates will be compared to published infection rates using episode rates per 1000 catheter days. These rates vary widely with infection rates ranging from 0.2 to 5.5 episodes per 1000 catheter days (Allon, 2008; Bakke, 2010; Lok et al, 2003; Maki, Kluger, & Crnich 2006; NKF, 2006). The National Healthcare Safety Network (NHSN) reported the pooled mean bacteremia infection rates in HD catheters used in outpatient HD as 1.4 episodes per 1000 catheter days (Kallen, Patel, & O'Grady, 2010).

Infection is a leading cause of mortality for patients with end stage renal disease (ESRD), with only cardiovascular disease resulting in a higher ESRD mortality (United States Renal Data System [USRDS], 2013). Only 52% of patients on HD are still alive three years after the start of dialysis (USRDS, 2013).

Increased infection risk may be caused by multiple factors, but vascular access type, poor patient hygiene, diabetes, hospitalization frequency, Staphylococcus aureus nasal carriage carriers, and immunosuppressants often increase the infection risk (Li et al, 2009; McCann & Moore, 2010; Vandecasteele, Boelaert, & DeVriese, 2009). In 2009, Vandecasteele et al. reported that the type of vascular access is the strongest predictor of infection risk, with catheters being the major culprit.


The RenalPro listserv ( search assisted the investigators in locating an outpatient HD center (London Health Science Centre) that had developed its own self-labeled "best practice" HD catheter care that incorporated no dressing and a prescribed shower routine (J.A. Lawrence, personal communication, July 2011). This center shared its protocol, exclusion criteria, and shower routine, which provided the initial study template. This best practice protocol started as a pilot study in 2008 and has been implemented at three incenter and nine satellite clinics (J.A. Lawrence, personal communication, July 2011). This program was initiated due to problems with gauze dressing adherence to the skin, catheter adhesive residue, exit site irritation, and crusting on the catheter that they hypothesized increased CRBSIs (J.A. Lawrence, personal communication, July 2011). They eliminated the dressing over the catheter exit site and instructed patients to shower following a prescribed procedure (Pember & Seiler, 2012). The exit sites and catheters appeared more pristine after this intervention, and all patients had positive comments. Their post-intervention CRBSI rate was 0.31 episodes per 1000 catheter days (J.A. Lawrence, personal communication, August 2012).

Study Design

This American Nephrology Nurses' Association (ANNA)-funded study was planned as a six-month prospective two-group randomized controlled trial (RCT). Participants were initially randomized into two groups (comparison and control group) using the computerized program GraphPad ( The comparison group was instructed in a prescribed showering technique (shower group), and the control group did not shower (non-shower group). The HD catheter exit site had no dressing coverage in both groups. The study protocol was approved by the Institutional Review Board (IRB) of a local hospital, and permission to conduct the study was obtained from a large dialysis organizations' research committee, the medical directors of the HD centers, and the ANNA Board of Directors. Informed consent was obtained before data collection began.

The catheter day denominator count began the day of study enrollment and continued until study closure, patient disenrollment, or death. The catheter day count continued during all hospitalizations throughout all the dialysis clinics. The catheter day denominator is the total population of this study for all the days they are in the study.

About a month after the study started, participants refused to be part of the study unless they were able to enroll in the showering arm of the study. The desire to openly shower was a very powerful inducement to enter the study, and the randomized control trial option had to be abandoned, allowing all on-going participants to select their showering/non-showering preference. An addendum reporting the change was submitted and approved by the IRB.

Study Population

The study population consisted of individuals with ESRD undergoing incenter HD using a tunneled cuffed catheter as the vascular access. The study was conducted in a total of nine HD centers located in rural and urban areas of the southwestern United States. Participants from nine clinics were recruited for the study. An estimated sample size of 78 subjects was needed to demonstrate statistical significance of alpha = 0.05 and power = 0.80 settings. The inclusion criteria included patients who were over 18 years of age, were Spanish- or English-speaking, had a well-healed (over one month since placement) HD catheter without sutures, used no antibiotics or antibiotic locks for catheter packing for four weeks prior to study entry, and were cognitively able to verbalize catheter care and able to "teach back" shower and catheter care. Additional criteria included patients who were physically able to perform their own catheter care, were not on immunosuppressants, did not sleep with pets, had no femoral or non-tunneled catheters, were without history in the past month of methicillin-resistant Staphylococcus aureus (MRSA) or vancomycin-resistant Enterococcus (VILE) CRBSIs, had no active malignancy receiving chemotherapy or radiation therapy, experienced no active skin condition involving the catheter site, and attended HD treatments regularly (missing less than one treatment per month over the previous three months).

All participants who met the inclusion criteria were approached by the PI to query their interest in the study. If they expressed an interest, the PI obtained informed consent. Participants had a set of screening blood cultures drawn after study consents were obtained. A negative blood culture screen was needed for study inclusion with the premise of not enrolling participants who had covert positive blood cultures.


During the study enrollment visit, the PI taught the participants the prescribed shower technique, and exit site and catheter care. Patients were allowed to shower at their own desired frequency. The participants were required to teach back the shower routine before they were provided the shower kit with supplies and written instructions (see Table 2). Mask use for exit site care in both groups was eliminated because the catheter exit site was continually left uncovered. The non-shower group was instructed by the PI on bathing options during the study enrollment visit. This allowed the participants to customize their bath to either a sit down tub bath, hand shower with no water near their catheter site, or a sponge bath. Participants in the non-shower group would not be getting the catheter site wet nor cleaning the catheter exit site, and would continue to leave the HD catheter exit site uncovered.

All exit sites continued to be assessed at every treatment by dialysis staff, with staff only cleaning exit sites in the non-shower group. Hub care was the same for both groups using the Chlorascrub[R] wipes (PDI) for "scrub the hub" care with both staff and patients wearing masks (see Tables 3 and 4). If the participant's HD catheter hubs were disconnected for any reason, an additional "scrub the hub" was done, including at dialysis completion.

All participants were advised to secure the catheter tips either in a bra or snug t-shirt to prevent catheter movement because there was no dressing over the catheter. Participants could secure the HD catheter tubing to the chest with tape if they chose, but were instructed not to apply tape to the caps. Lists were posted in each clinic designating the participants as either belonging to the shower or non-shower group. Hand hygiene was emphasized for both staff and participants.

The dialysis clinics and participants were provided with ongoing supplies for the six-month study at no cost. All participants who completed the study were compensated $25.00 at study closure. The dialysis clinics were thanked for their participation with a $25.00 gift.

There was no change in the routine of the dialysis clinics except for the exit site and hub care. The dialysis clinic staff and providers could easily assess the HD catheter exit sites. Due to the distances between the dialysis clinics, it was not feasible to routinely observe staff adherence to study protocol, so the investigators obtained verbal confirmation of adherence that staff and participants reported following the protocol as instructed. The exit site and no-dressing use were monitored by the nephrology providers during weekly dialysis clinic visits. Hub care was not formally monitored by study personnel who relied on a verbal report from the staff at study closure.


Of the 74 individuals invited to participate, 27 (36%) elected not to participate citing two major reasons: commitment to previous instruction by their nurses to never shower and to avoid touching their dressings; and concern with doing anything different than their usual routine because of a perceived increase in infection risk.

Two clinics were 300 miles away from the workplace of the PI. After the distant rural clinic staff members were oriented to the study, there was a lack of communication between staff and the PI, and the sites were deemed too distant for proper follow up to ensure fidelity to the research design. Therefore, to provide appropriate participant safety, the two distant rural clinics were closed as study sites two months after study initiation. Participants were informed via the consent form that if there were issues, they may be withdrawn from the study. Closing these two sites resulted in the early withdrawal of seven patients from the study. Their data were included in the statistical analysis with their length of study days based on their actual days of study participation for their catheter day denomination count.

Three participants were enrolled in the study, but their blood culture kits were not delivered for testing per protocol until after the participants left the clinic. The staff at the clinic did not draw the blood cultures at the participants' next dialysis treatment and after the PI discovered the missed blood cultures, the participants had already been in the study for a week, which meant results would be unreliable. As a result, their blood cultures were not drawn; the PI decided to allow them to stay in the because since they had no symptoms of bacteremia.

A total of 40 participants from nine dialysis clinics were enrolled and remained in the study with study days per participant ranging from 25 to 200 days. Thirty-one participants were in the shower group, and nine participants were in the non-shower group. The participation goal of 78 participants was not reached due to a lack of study personnel and participant attrition. See Figure 1 for a diagram of study flow.

Baseline characteristics of participants in total and by group (shower and non-shower) are detailed in Table 3. The data analysis included the 40 participants who completed the study and were followed up to six months. The most common reasons for less than six months of study follow up included transition to a mature fistula or graft (13%), rural clinic closure (10%), change in patient residence (2.5%), patient self-withdrawal (2.5%), and one patient death (2.5%). This study monitored for CRIs for a total of 5165 catheter days. The ratio of participants in the shower vs. non shower groups was 74% (n = 31) in the shower group and 26% (n = 9) in the non-shower group.

Descriptive statistics (frequency counts, percentages, and means) were used to describe the variables as appropriate. Fisher's exact tests were conducted to identify significant differences between the groups, but no differences were found. The statistical analysis demonstrated no significance in the outcomes due to the inadequate subject enrollment, limited number of infections, and the unanticipated early discontinuation of randomization.

Primary Outcomes--Infection

Definitions are shown in Table 1. Infection results are shown in Table 5.

Symptomatic bacteremia. Of the 40 participants who completed the study, two (5%) participants had symptomatic bacteremia (3.2% [n = 1] in the shower group, 11.1% [n = 1] in the non-shower group, which was not statistically significant (p = 0.404). Bacteremia occurred in these two patients at day 75 and day 136 from study enrollment date. Both participants had urinary tract infections; Escherichia coli (E-coli) was the pathogen in each of the participant's blood and urine cultures, and was treated with conventional antibiotic therapy. One participant had the HD catheter removed, resuming use of the arteriovenous fistula. The other participant had a foot ulcer that responded to antibiotics, and the HD catheter remained in place. So the participant continued to participate in the study. One participant was in the shower group, and the other participant was in the non-shower group. Each HD catheter was less than three months old.

Tunnel site infections. Of the 40 participants who completed the study, three (7.5%) participants had tunnel site infections (9.7% In = 3] in the shower group, 0% in the nonshower group), which was not statistically significant. The tunnel infections occurred on day 25, day 75, and day 180 from study enrollment. One culture result was negative and the other two cultures grew MSSA. The culture results were 1) MSSA-positive wound culture tunnel secretions with negative blood culture; 2) MSSA-positive blood cultures with negative wound culture; and 3) all negative cultures. All participants required HD catheter removal; one participant died, one participant transferred to peritoneal dialysis (PD), and one participant had a new HD catheter placed and resumed incenter dialysis. Two catheters were less than six months old, and both grew MSSA; the third catheter was eight months old with a negative culture.

Exit site infections. Of the 40 participants who completed this study, no participants were found to have exit site infections.

Catheter-related bloodstream infections. Of the 40 participants who completed this study, two participants (5%) had CRBSIs (6.5% [n = 2] in the shower group, 0% in the non-shower group), which was not statistically significant. The incidents of CRBSIs occurred on day 25 and day 185 from study enrollment. One participant had a co-existing bacteremia and tunnel infection (also reported above), which grew MSSA from the blood cultures. The other participant had multiple negative blood cultures, but received IV antibiotics and HD catheter removal due to persistent fever and chills, which occurred only when the participant was on dialysis. This participant's symptoms resolved after completing the IV antibiotics, HD catheter removal, and delayed HD catheter replacement. One catheter was less than six months old, and the second catheter was over a year old.

Calculated infection rates per 1000 catheter days. The non-shower group for this six-month study had zero CRBS1 episodes per 1000 catheter days, while the shower group had a rate of 0.51 CRBSI episodes per 1000 catheter days. The CRBSI episode rate for both groups combined was 0.39 per 1000 catheter days. This rate is in the low range of the published results of 0.2 to 5.5 episodes per 1000 catheter days (Allon, 2008; Bakke, 2010; Lok et al, 2003).

The tunnel infection rate was 0.58 episodes per 1000 catheter days, all in the shower group. No comparable data were located in the literature for a comparison. The exit site infection rate for this study was 0 per 1000 catheter days compared to 1.02 episodes per 1000 catheter days found by Lok et al. (2003) and 0.6 episodes per 1000 catheter days from a meta-analysis by James et al. (2008).

Secondary Outcomes--Hospitalization Due To Infection

Of the 40 participants who completed this study, two participants (50/0) required hospitalization related to infection (3.2% [n = 1] in the shower group, 11.1% [n = 1] in the non-shower group). All other care was provided on an outpatient basis with antibiotic courses and HD catheter changes (if indicated) as ordered by the participants' nephrology providers with resolution of the infection without sequalae. Allon (2004) noted a 10% hospitalization rate in patients on dialysis with CRBSIs due to severe sepsis or metastatic infections, while this study had a 5% hospitalization rate due to sepsis throughout this six-month study.

Secondary Outcomes--Mortality

There were four deaths (10%) in the total sample (12.9% [n = 4] in the shower arm and 0 deaths in the non-shower arm, not statistically significant (p= 0.557). Three of these deaths occurred following self-withdrawal from dialysis. The reasons for dialysis withdrawal included cancer (2.5% [n = 1]), chronic pain (2.5% [n = 1), and failure to thrive (2.5% [n = 1]), all conditions unrelated to the study. None of the participants who self-withdrew from dialysis had a CRBSI.

One participant in the shower-arm group was hospitalized with sepsis within 25 days of enrolling in the study with a tunnel infection and CRBSI. The participant was recovering from the infection as demonstrated by fever resolution, negative follow-up blood cultures, and documented clinical signs of improvement, but experienced a cardiovascular event and expired during unsuccessful cardiopulmonary resuscitation. This participant had a very low serum albumin (not an exclusion criteria) and was a smoker. It was postulated that the tunnel tract was not totally healed with these underlying risk factors. This event led to a change in study inclusion criteria, a new definition of well-healed HD catheter tunnel tract, and incorporating a tug test into ongoing catheter evaluations (See Table 1 for the definition of tug test.)

Ramanathan et al. (2007) showed a risk for higher mortality in hospitalized patients with CRBSI who also had low serum albumin. One single-center study of patients on HD with S. aureus bacteremia had in-hospital mortality of 9.5% and 12-week mortality up to 19% (Engemann et al., 2005). These S. aureus bacteremia mortality rates were re-confirmed in a multi-site study by Li et al. (2009). Mokrzycki and Lok (2010) found that after any episode of sepsis, adverse cardiovascular events also increase two-fold.

This study had a 10% mortality rate in the six-month time period, the majority of the mortality (three of four patients) was due to self-withdrawal from dialysis. The mortality rate in the United States of patients in their first year of hemodialysis is 26%, with the unadjusted mortality rate for prevalent patients with ESRD at 245 deaths per 1000 patient years (USRDS, 2012).


The one episode of positive CRBSI blood cultures was due to a S. aureus organism (MSSA). This CRBSI required hospitalization, and one of the non-catheter-related bacteremias also required hospitalization. The second CRBSI was never confirmed by blood cultures, but the participant's fever resolved after catheter removal, and no other source of infection was ever identified. All the other infections were treated on an outpatient basis in the patients' dialysis clinics.

The two bacteremia infections were presumed to not be catheter-related because both had urinary tract infections, and one also had a foot ulcer. Both had positive E-coli organisms in their blood and urine. One participant had his catheter removed because his arteriovenous fistula was mature at that time, and the other participant finished his antibiotic course without removing the HD catheter.

All three tunnel infections occurred in participants who were showering. The first participant had fever and co-existing tunnel infection and CRBSI that resulted in hospitalization. The second episode of tunnel infection occurred in a participant who was preparing to transfer to peritoneal dialysis (PD) and had the PD catheter in place. The third episode of tunnel infection was atypical. The infection started as swelling, redness, and pain at the mastoid area that over a few days descended into the tunnel tract, evolving into a tunnel infection. This HD catheter was removed, IV antibiotics administered, and a new HD catheter was placed.

The limitations of this study included the 1) enrollment not achieving the sample size of 78 participants for the study design, which lowered the power of the statistical analysis; 2) inability to continue with randomization of the shower versus non-shower groups early in the study; 3) inability to monitor staff and patients for breaches in protocol; and 4) the study design did not allow assessment of the impact of no dressing.

Implications for Nephrology Nursing

Nephrology nurses strive to provide the best care to individuals undergoing HD, and one important aspect is to reduce the frequency of CRBSIs. Considering the number of people with HD catheters, an evidence-based protocol is essential to achieve this goal. This study demonstrated that allowing patients to shower and not cover the HD catheter with a dressing may be a low-cost alternative to current practice that might be associated with low CRBSI and tunnel infection rates. A research study from the University of Michigan projected each CRBSI as costing about $45,000 (Pronovost et al., 2010). The difference in the infection rate in this study (0.39 episodes per 1000 days) and the NHSN CRBSI rate (1.4 episodes) is a difference of 1 episode per 1000 catheter days, which is a savings of $45,000 per 1000 catheter days along with the decreased morbidities associated with CRBSIs.

One very important aspect of this study is the empowering effect verbalized by most of the participants, which was an unexpected positive finding (Kear el al., 2013). Having patients do more of their care has been shown in other areas of chronic care management to improve health outcomes (Anderson & Funnell, 2010). Further research exploring interventions that involve patients in HD catheter care and exploring patients' catheter care practices in the outpatient setting is needed.

Catheter hub care is critically important, and it must include a "scrub the hub" technique each time the hub connections are accessed and integrity of the internal catheter space is disturbed (dialysis initiation, termination of dialysis). No conclusions can be drawn related to patients who wish to shower but do not want to do their exit site care after the shower. This particular scenario was beyond the scope of this study. There is no substitute for hand hygiene by staff and patients, and this must be carefully monitored. Patients need to secure their HD catheter, by either tucking the HD catheter tips into a bra or wearing a snug t-shirt, especially at night.

To answer the research questions for this study: The no-dressing coverage over the HD catheter achieved a low CRBSI rate (0.39 per 1000 catheter days compared to published data of 0.2 to 5.5 per 1000 catheter days), tunnel infections (0.58 per 1000 catheter days compared to no published data), and exit site infections (0 per 1000 catheter days compared to 1.02 per 1000 catheter days). The group of participants who showered had a higher incidence rate of CRBSIs and tunnel infections compared to the group of participants that did not shower. The answers to these research questions must be analyzed carefully because this was a small study, and just one additional patient with a positive finding in either group would have altered the results.


It is hoped this study will stimulate more researchers to continue investigating best practices in HD catheter care because this area of research has been limited, and participants verbalized their desire to continue to shower. Due to attrition, limited study help, and few infections, the study did not achieve statistical significance. The smaller sample size lowered the statistical power of the study.

This study replicated with a larger sample size would strengthen the evidence. Several positive patient outcomes require further investigation. The no-dressing coverage over HD catheter resulted in zero exit site infections. Neither patients nor staff wore masks during exit site care. From the participants' point of view, the prescribed showering routine and no tape were the two most positive aspects of this study, and all participants verbalized a desire to stay with this routine. The showering group had the higher infection rate (bacteremia, CRBSI, tunnel infections) but still achieved a low incidence level of infections compared to other published results (Allon, 2008; Bakke, 2010; Lok et al, 2003; Maki et al., 2006). The researchers believe the desirability of showering resulted in the strong patient adherence to the protocol steps, which was confirmed verbally by all participants during the closing study interview (Kear et al., 2013).

The evidence would be strengthened with replication by another study validating the definition of a well-healed HD catheter. The catheter tug test was initiated as a tool to help evaluate well-healed HD catheters and consists of a gentle tug on the catheter to monitor for catheter downward movement or cuff exposure. Once replicated on a larger scale, this pilot study has the potential to add to the evidence for HD catheter care, thus allowing nephrology nurses to base their HD catheter patient education and care more on science than tradition.

Acknowledgments; The authors extend their thanks to Dr. Alexander Kallen from the Centers for Disease Control and Prevention for his assistance and recommendations; to Renal Medicine Associates (Dr. Kirby Gabrys, Dr. Fidel Barrantes, Dr. Jayant Kumar, Dr. Sonata Kundeling, Dr. Biju Cherian, Dr. Chudi Adi, Dr. Gustavo Espino, Dr. Leonard Romero) for approving the protocol to be implemented at their dialysis clinics; and to the London Health Science Centre for sharing their experiences and protocoL

Source of Funding: This study was supported by a grant made available by the American Nephrology Nurses' Association (ANNA). Findings of the study do not necessarily reflect the opinions of ANNA. The views expressed herein are those of the author(s), and no official endorsement by ANNA is intended or should be inferred.

Note: Authors' biographical statements, acknowledgments, and source of funding, and can be found on the following page.

Editor's Note: Please refer to the article by Julie Ann Lawrence, Suzanne Siler, Barbara Wilson, and Lori Harwood, "Shower and No-Dressing Technique for Tunneled Central Venous Hemodialysis Catheters: A Quality Improvement Initiative," which appears on pages 67-72 of this issue, for additional information on this topic

Statement of Disclosure: The authors reported no actual or potential conflict of interest in relation to this continuing nursing education activity.

Note: Additional statements of disclosure and instructions for CNE evaluation can be found on page 65.


Allon, M. (2004). Review: Dialysis catheter-related bacteremia: Treatment and prophylaxis. American Journal of Kidney Diseases, 44(5), 779-791.

Allon, M. (2008). Editorial: Prophylaxis against dialysis catheter-related bacteremia: A glimmer of hope. American Journal of Kidney Diseases, 5(2), 165-168.

Anderson, R.M., & Funnell, M.M. (2010). Patient empowerment: Myths & misconceptions. Patient Education and Counseling, 79(3), 277-282.

Bakke, C.K. (2010). Clinical and cost effectiveness of guidelines to prevent intravascular catheter-related infections in patients on hemodialysis. Nephrology Nursing Journal, 37(6), 601-615.

Beathard, G.A., & Urbanes, A. (2008). Infection associated with tunneled hemodialysis catheters. Seminars in Dialysis, 27(6), 528-535.

Betjes, M.G. (2011). Prevention of catheter-related bloodstream infection in patients on hemodialysis. Nature Reviews Nephrology, 7(5), 257-265.

Centers for Disease Control and Prevention (CDC). (2013). National Health Safety Network (NHSN) July 2073 CDC/NHSN Protocol clarification. Retrieved from

Dinwiddie, L.C., & Bhola, C. (2010). Hemodialysis catheter care: Current recommendations for nursing practice in North America. Nephrology Nursing Journal, 37(5), 507-521.

Dunbar Ivy, D., Calderbank, M., Wagner, B.D., Dolan, S., Nyquist, A.C., Wade, M., ... Doran, A.K. (2009). Closed-hub systems with protected connections and the reduction of risk of catheter-related bloodstream infection in pediatric patients receiving intravenous prostanoid therapy for pulmonary hypertension. Infection Control and Hospital Epidemiology, 30(9), 823-829.

Engemann, J.J., Friedman, J.Y., Reed, S.D., Grifilths, R.I., Szczech, L.A., Kaye, K.S., ... Fowler, V.G. (2005). Clinical outcomes and costs due to Staphylococcus aureus bacteremia among patients receiving long-term hemodialysis. Infection Control Hospital Epidemiology, 26(6), 534-539.

Frasca, D., Dahyot-Fizelier, C., & Mimoz, O. (2010). Prevention of central venous catheter-related infection in the intensive care unit. Critical Care 2070, 74(2), 212-220.

James, M.T., Conley, J., Tonnelli, M., Manns, BJ., MacRae, J., & Hemmelgam, B.R. (2008). Metaanalysis: Antibiotics for prophylaxis again hemodialysis catheter-related infections. Annals of Internal Medicine, 748(8), 596-605.

Kallen, A.J., Arduino, M.J., & Patel, ER. (2010). Preventing infections in patients undergoing hemodialysis: Current infection control issues in hemodialysis. Expert Reviews--Anti-Infective Therapy, 8(6), 643-655.

Kallen, A.J., Patel, P.R., & O'Grady, N.E (2010). Preventing catheter-related bloodstream infections outside the intensive care unit: Expanding prevention to new settings. Healthcare Epidemiology, 57(3), 335-341.

Kear, T., Evans, E., Hain, D., Schrauf, C., & Dork, L. (2013). Patients' perceptions of hemodialysis catheter care practices at home before and after eliminating a protective dressing and implementing a showering protocol. Journal of Infection Prevention, 74(6), 208-212. doi: 10.1177/1757177413495908

Li, Y., Friedman, J.Y., O'Neal, B.F., Hohenboken, M.J., Griffiths, R.I., Stryjewski, M.E., ... Reed, S.D. (2009). Outcomes of Staphylococcus aureus infection in hemodialysis-dependent patients. Clinical Journal American Society of Nephrology, 4(2), 428-434.

Lok, C.E., Stanley, K.E., Hux, J.E., Richardson, R., Tobe, S.W., & Conly, j. (2003). Hemodialysis infection prevention with polysporin ointment. Journal of American Society of Nephrology, 14(1), 169-179.

Maki, D.G., Kluger, D.M., & Cmich, C.J. (2006). The risk of blood stream infections in adults with different intravascular devices: A systematic review of 200 published prospective studies. Mayo Clinic Proceedings, 87(9), 1159-1171.

McCann, M., & Moore, A.E. (2010). Interventions for preventing infectious complications in haemodialysis patients with central venous catheters. Cochrane Database Systematic Review, 7, CD006894. doi:10.1002/14651858

Mokrzycki, M.H., & Lok, C.E. (2010). Advances in hemodialysis catheter-related bacteremia. Touch Briefings, 5(1), 84-88.

National Kidney Foundation (NKF). (2006). KDOQI clinical practice guidelines and clinical practice recommendations for 2006 updates: Hemodialysis adequacy, peritoneal dialysis adequacy and vascular access. American Journal of Kidney Diseases, 48(Suppl. 1), S1-S322. Retrieved from

O'Grady, N.E, Alexander, M., Dellinger, E.E, Gerberding, J.L., Heard, S.O., Maki, D.G., ... Weinstein, R., (2002). Guidelines for the prevention of intravascular catheter-related infections. American Journal of Infection Control, 30(8), 476-489.

O'Grady, N.E, Alexander, M., Bums, L.A., Dellinger, E.E, Garland, J., Heard, S.O., ... Healthcare Infection Control Practices Advisory Committee (HICPAC) (2011). Guidelines for the prevention of intravascular catheter-related infections. Clinical Infectious Diseases, 52(9), e162-e193. Retrieved from

Olson, K., Rennie, R.E, Hanson, J., Ryan, M., Gilpin, J., Falsetti, M., ... Gaudet, S. (2004). Evaluation of a no-dressing intervention for tunneled central venous catheter exit sites. Journal of Infusion Nursing, 27, 37-44.

Pember, A., & Seiler, S. (2012) Shower and no dressing technique for tunneled central venous hemodialysis catheters: 2012--An update. Canadian Association of Nephrology Nurses and Technicians, 22(2), 16.

Petrosino, B., Becker, H., & Christian, B. (1988). Infection rates in central venous catheter dressings. Oncology Nursing Forum, 75(6), 709-717.

Pronovost, E, Goeschel, C.A., Colantuoni, E., Watson, S., Lubomski, L.H., Berenholtz, S.M .... Needham, D. (2010). Sustaining reductions in catheter related bloodstream infections in Michigan intensive care units: Observational study. British Medical Journal, 340, c309.

Ramanathan, V., Chiu, E.J., Thomas, J.T., Khan, A., Dolson, G.M., & Darouiche, R.O. (2007). Healthcare costs associated with hemodialysis catheter-related infections: A single-center experience. Infection Control and Hospital Epidemiology, 28(5), 606-609.

U.S. Renal Data System (USRDS). (2012). USRDS 2012 annual data report: Atlas of chronic kidney disease and end-stage renal disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.

U.S. Renal Data System (USRDS). (2013). USRDS 2013 annual data report: Atlas of chronic kidney disease and end-stage renal disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.

Vandecasteele, SJ., Boelaert, J.R., & DeVriese, A.S. (2009). Staphylococcus aureus infections in hemodialysis: What a nephrologist should know. Clinical Journal of American Society of Nephrology, 4(8), 1388-1400.

Wilcox, T.A. (2009). Catheter--Related bloodstream infections. Seminars in Interventional Radiology, 26(2), 139-143.

Evans, E.C., Hain, D., Kear, T.M., Dork, L.A., & Schrauf, C. (2014). Hemodialysis catheter outcomes pilot study: No dressing coverage with prescribed showering. Nephrology Nursing Journal, 41(1), 53-64, 72.

Elizabeth Evans, DNP, is a Nephrology Nurse Practitioner, Renal Medicine Associates, Albuquerque, NM; ANNA Research Chair 2011-2013; and Member and Secretary/Treasurer of ANNA's Nuevo Mexico Chapter. She may be contacted directly via email at

Debra Hain, PhD, APRN, ANP, GNP-BC, is an Assistant Professor of Nursing, Florida Atlantic University, Boca Raton, FL; a Nurse Practitioner, Department of Nephrology, Cleveland Clinic Florida, Weston, FL; 2013 Chair of the ANNA Research Committee; an ANNA Web Fellow; an ANNA Representative on the Congress on Nursing Practice and Economics; and a member of ANNA's South Florida Flamingo Chapter.

Tamara M. Kear, PhD, RN, CNN, is an Assistant Professor of Nursing, Villanova University, Villanova, PA; a Nephrology Nurse, Fresenius Medical Care-North America, Waltham, MA; a member of the ANNA Research Committee 2012-2013; the ANNA representative for the American Nurses Association (ANA) Working Group on "Racism and Diversity;" a member of the ANNA Competency Based Governance Task Force; a member of the Nephrology Nursing Journal Editorial Board; and a member of ANNA's Keystone Chapter.

Leslie Dork DNP(c), is a Nephrology Nurse Practitioner, Renal Medicine Associates, Albuquerque, NM; a member of the ANNA Research Committee 2011-2013; and a member of ANNA's Nuevo Mexico Chapter.

Christine Schrauf PhD, RN, MBA, is an Associate Professor of Nursing, Elms College, Chicopee, MA; a member of the ANNA Research Committee," and a member of ANNA's Colonial Chapter.

Table 1

Blood culture and exit site culture technique for fever
protocol--Renal Medicine Associates (RMA) fever guidelines
(temperatures greater than 99.9 degrees F) will be followed at
participating clinics. Draw two sets of blood cultures (two aerobic
and two anaerobic) of 10 mL of blood from venous line and arterial
line 15 to 30 minutes apart for each set. Notify provider of fever
(and if any potential positive source: lungs, catheter site,
urinary tract infection, etc). If the access site is red and/or has
drainage/culture drainage, notify provider (RMA, personal
communication, 2010).

Bloodstream infection (BSI): Secondary to site-specific
infection--Suspected of having an infection, blood and a
site-specific organism are collected for a culture, and both are
positive for at least one matching organism. If the site-specific
culture is an element used to meet the infection site criterion,
then BSI is considered secondary to that site-specific infection
(CDC, 2013).

Catheter days denominator--Catheter days will be a direct count and
not an estimate for each catheter to fully account for exposure
time in the study. The catheter count denominator begins the date
of study enrollment throughout hospitalizations and end the day of
study closure, patient disenrollment, or death. These data were
checked by computer calculation.

Catheter-related bacteremia--Has recognized pathogen cultured from
one or more blood cultures, and organism cultured from blood is not
related to an infection at another site (CDC, 2013).

Exit site infection--Inflammation confined to the area surrounding
the catheter exit site, not extending superiorly beyond the cuff
because the catheter is tunneled with the exit site having
exudates. The culture of catheter exudate has been confirmed to be
positive (NKF, 2006).

Tunnel infection. The catheter tunnel superior to the cuff is
inflamed, painful, and may have drainage through the exit site that
is culture positive (NKF, 2006). There is presence of tenderness,
erythema, or induration over 2 cm from the exit site along the
subcutaneous tract of the tunnel with or without evidence of
bloodstream infection (Wilcox, 2009).

Well-healed catheter--RMA definition: Well-healed tunneled catheter
is over four weeks since placement with no sutures, redness,
tenderness, or drainage present at the exit site. The catheter will
be gently tugged to assess for catheter movement--if the catheter
moves, then the catheter will not be deemed well healed until three
months after placement. If the patient is a smoker with serum
albumin less than 3.2 mg/dL, the catheter will not be deemed well
healed until three months after placement (RMA, personal
communication, 2011).

Table 2
Patient Procedures--Shower Technique for HD Catheter Care: Patient

Patient Equipment

Shower Kit (two bars of unscented Dove[R] soap, Glad Press 'n
Seal[R] roll, scissors, one box of Chlorascrub[R] wipes, sleeve of
bulk unsterile 3x3s--all in a plastic container with lid). Patient
to use own clean towel for drying.


1. Make sure the catheter caps are secure before showering.

2. Cut off a section of Press 'n Seal and wrap securely around the
catheter hubs.

3. Gather all equipment before entering the shower.

4. In shower, wash and rinse face, hair first, then wash the
remainder of the body. Wash catheter last.

5. Prior to washing the catheter, wash your hands with soap and
water again. Gently wash the skin around the catheter with Dove
soap with your clean hands. Only use non-perfumed soap.

6. Rinse well and do not wash this area again. Once out of the
shower, gently pat the skin around the catheter with a clean gauze
dressing. Then dry the rest of the body with a towel.

7. Remove the Press 'n Seal wrap from the catheter hubs. Do not tug
on the catheter.

8. Open the Chlorascrub wipe. Cleanse skin around the catheter
using friction to clean side to side. Turn the pad over, and then
clean up and down using friction.

9. Pad the catheter tips with gauze or tape the catheter to your
skin per your preference.

10. Be sure to keep the catheter secure by wearing a bra and
tucking the ends into the bra or wear a snug t-shirt to help
prevent tugging on the catheter.

Table 3
Patient Procedures--Hub Care Procedure:
Dialysis Clinic Staff Instructions

Use this procedure anytime the hemodialysis catheter is disconnected
(i.e. beginning or ending dialysis or temporarily coming off dialysis
for a bathroom break or reversing lines).


Chlorhexidine wipes, sterile 3x3 gauze, usual syringes, and
heparin/normal saline for packing or flushing.

Connection Procedure

1. Wash hands for 15 seconds with soap and water or apply alcohol gel.
Allow to dry.

2. Patient and staff both apply mask. Apply a clean Chux[R] to the
clean chair-side table.

3. Put all syringes and supplies on chux and be sure to keep all
syringes in packet to keep sterile.

4. Apply non-sterile gloves. Be sure catheter is clamped.

5. Place a sterile 3x3 gauze under the hubs. Remove the caps.

6. Open chlorhexidine wipe; scrub the Luer lock threads and hubs in a
back and forth vigorous movement for 15 seconds each. Repeat technique
with second wipe, scrubbing Luer lock threads and hubs. Allow to dry
thoroughly. Do not wipe or blot.

7. Before accessing any injection ports, scrub vigorously all ports
with alcohol wipe for 15 seconds to reduce contamination.

8. Clinic staff must wash hands with soap or use alcohol-based
waterless cleanser:

  * Before and after palpating catheter insertion site.

  * Before and after accessing, repairing, or dressing an intravascular
    catheter; this includes associated components, such as
    administration sets and access ports.

  * The use of gloves does not obviate the need for hand hygiene.

  * Educate patients, caregivers, and family about the importance of
    hand hygiene and ask that they remind caregivers to clean their

9. If the patient has to go to the bathroom or lines need to be
reversed, disconnect and scrub Luer locks and hubs vigorously with one
chlorhexidine wipe for 15 seconds and allow to air dry. When the
patient returns from bathroom, scrub Luer lock threads and hubs
vigorously with one chlorhexidine wipe for 15 seconds, allow to dry,
and then reconnect dialysis. When patients return from the bathroom,
have patient and staff apply alcohol-based hand gel before staff member
applies gloves. Both apply mask.

10. Use caution to be sure that no hub connector is dropped after
scrubbing with chlorhexidine; chest gauze will help keep site clean.

11. If syringes or needles slip out of the sterile package, discard
and obtain new sterile supplies.

12. After dialysis, disconnect and scrub Luer lock threads and hubs
again with one separate chlorhexidine wipe for 15 seconds each. Allow
to dry thoroughly and do not wipe or blot.

Table 4
Patient Procedures--Exit Site Care Procedure:
Dialysis Clinic Staff Instructions

1. Shower arm patients: No exit site cleaning is required. Patients
will be doing their cleaning of the site after their shower. Just
confirm they showered and cleaned the site with chlorhexidine wipe.

2. No mask is required by patient or staff to assess the exit site.
Exit site must be assessed by the staff every dialysis visit.
Cleanse hands and apply non-sterile gloves prior to touching and
assessing exit site. No dressing is applied.

3. Non-shower arm patients: No mask is required by patient or staff
for exit site care. Cleanse hands and apply non-sterile gloves. Cleanse
exit site with one chlorhexidine wipe using a vigorous back and forth
scrubbing movement, starting at the catheter exit site and working
outwards while being sure to clean under the catheter and also cleaning
the catheter. No dressing is applied.

4. Exit site assessments: Observe for any unusual skin problems,
tenderness, cuff exposure, or drainage. With gloved hands, touch exit
site and express around catheter checking for the above.

Table 5
Patient Characteristics and Outcomes Based on Shower/Non-Shower Group

                                       Shower Group n(%)

Patients                                   31 (78.0)

  Caucasian                                11 (27.5)
  Native American                           5 (12.5)
  African American                          3 (7.5)
  Hispanic                                 12 (30.0)

  Male                                     12 (30.0)
  Female                                   19 (47.5)

Duration on Dialysis                  1 month to 12 years;
                                        Mean 36.5 months

Etiology of ESRD
  Diabetes                                 18 (45.0)
  Unknown                                   1 (2.5)
  Renal cancer                              1 (2.5)
  Hypertension                              3 (7.5)
  Glomerulonephritis                        2 (5.0)
  Obstructive uropathy                      1 (2.5)
  FSGS                                      2 (5.0)
  ADPKD                                     1 (2.5)
  Membranous nephropathy                    1 (2.5)
  Acute kidney injury                       1 (2.5)

Patient Age                              Mean 64 years

Catheter Days in Study                        3935

Catheter Age                          1 month to 5 years;
                                         Mean 9 months;
                                        Median 8 months

Hospitalizations                            1 (2.5)

Mortality                                   4 (10.0)
  Causes                   #1 patient: hospitalized and recovering
                           from MSSA CRBSI
                           #2 patient: self-withdrew from dialysis
                           due to cancer
                           #3 patient: self-withdrew from dialysis
                           due to failure to thrive
                           #4 patient: self-withdrew from dialysis
                           due to chronic pain and s/p hip fracture

Tunnel Infections                           3 (7.5)

CRBSI                                       2 (5.0)
                           #1 patient: MSSA blood culture (BC)
                           #2 patient: Negative cultures

Bacteremia                                  1 (2.5)
  Organism                 #1 patient: E-coli

Asymptomatic Bacteremia                     2 (5.0)
  Organism                 #1 patient: Staphylococcus capitis ssp
                           ureolyticus--second BC recheck
                           (catheter only due to no peripheral veins)
                           same organism - IV antibiotics and
                           catheter change.
                           #2 patient: Staphyloccocus epidermidis-
                           second BC recheck with peripheral veins
                           only--no organism, no antibiotics

Exit Site Infections       0 - No organisms/infections

                             Non-Shower Group n(%)     Total n(%)

Patients                            9 (22.0)           40 (100)

  Caucasian                         2 (5.0)            13 (32.5)
  Native American                   2 (5.0)             7 (17.5)
  African American                  0 (0)               3 (7.5)
  Hispanic                          5 (12.5)           17 (42.5)

  Male                              6 (15.0)           18 (45.0)
  Female                            3 (7.5)            22 (55.0)

Duration on Dialysis         5 months to 7 years;
                                Mean 42 months

Etiology of ESRD
  Diabetes                          4 (10.0)           22 (55.0)
  Unknown                             0                 1 (2.5)
  Renal cancer                        0                 1 (2.5)
  Hypertension                      2 (5.0)             5 (12.5)
  Glomerulonephritis                  0                 2 (5.0)
  Obstructive uropathy                0                 1 (2.5)
  FSGS                              1 (2.5)             3 (7.5)
  ADPKD                               0                 1 (2.5)
  Membranous nephropathy              0                 1 (2.5)
  Acute kidney injury               2 (5.0)             3 (7.5)

Patient Age                      Mean 63 years

Catheter Days in Study               1230                5165

Catheter Age                  1 month to 1 year;
                                Mean 9.6 months
                               Median 10 months

Hospitalizations                    1 (2.5)             2 (5.0)

Mortality                                               4 (10.0)

Tunnel Infections                      0                3 (7.5)

CRBSI                                  0                2 (5.0)

Bacteremia                          1 (2.5)             2 (5.0)
  Organism                 #2 patient: E-coli

Asymptomatic Bacteremia             1 (2.5)
  Organism                 #3 patient:
                           Staphylococcus               3 (7.5)
                           hominis with 2nd
                           BC--Gemella Morbillorum
                           from aerobic BC only
                           (catheter only). 3rd
                           BC--using only peripheral
                           veins--negative BC - IV
                           antibiotics stopped.

Exit Site Infections       0 - No organisms/               0
COPYRIGHT 2014 Jannetti Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:CNE: Continuing Nursing Education
Author:Evans, Elizabeth C.; Hain, Debra; Kear, Tamara M.; Dork, Leslie A.; Schrauf, Christine
Publication:Nephrology Nursing Journal
Article Type:Clinical report
Date:Jan 1, 2014
Previous Article:Error recovery by dialysis technicians.
Next Article:Shower and no-dressing technique for tunneled central venous hemodialysis catheters: a quality improvement initiative.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters