Bridging the gap: introduction of an antimicrobial peripherally inserted central catheter (PICC) in response to high PICC central line-associated bloodstream infection incidence*

HIGHLIGHTS

• Gap analysis identified an opportunity for introduction of antimicrobial PICCs to a large community hospital with elevated CLABSI incidence
• Combined with continued compliance with basic prevention strategies (ie use of a central line insertion checklist/insertion bundle) and optimization of device selection and lumen justification, the introduction of an antimicrobial antithrombogenic PICC was associated with a significant reduction in CLABSI

Central venous catheters represent lifesaving devices for patients with critical vascular access needs. Peripherally inserted central catheters(PICCs)are a common choice for patients requiring longer courses of therapy, particularly with infusates such as vasopressors, parenteral nutrition, and pH considerations. PICCs are often placed by specially trained nursing (or respiratory therapist) teams at the patient's bedside, using tip navigation technology that can eliminate the need for chest X-ray confirmation before use in many organizations.

In 2016, as part of the organization's annual infection control risk assessment, a gap analysis was performed regarding central line-associated bloodstream infection (CLABSI) reduction activities. Fifty percent of CLABSIs occurred in patients with PICCs, which continued through the next year before implementation of the new device. The guidelines and standards reviewed were all aligned in favor of antimicrobial (AM) device use when CLABSI rates remain elevated despite adherence to central line insertion bundle and other prevention strategies, as shown in Table 1.^1,2,3,4 A recent published systematic review and meta-analysis also supports reduction in CLABSI incidence with the use of AM devices.⁵ Previous studies have reviewed performance in acute care as well as long-term acute care settings, although not all have demonstrated decrease in infections.^6,7,8 In our gap analysis, the only other items beyond AM catheters that were noted as significantly discrepant from the reviewed recommendations were the team's overreliance on triple- and double-lumen catheters (single-lumen devices were not used at the time) and the absence of a trained observer for up to 50% of PICC placements. The nurse-led vascular access team (VAT) presented a proposal through Shared Governance to pilot and ultimately adopt an AM/antithrombogenic (AT) catheter (Arrowg+ard Blue Advance™ PICC). Additionally, lumen justification criteria were introduced into order sets, and the team added single-lumen devices to their available stock.

The VAT is comprised of 4 full-time nurses and 2 part-time nurses covering 2 hospital campuses. They work 10-hour shifts Monday through Friday, generally staffed with 2 nurses at the larger facility and 1 (rarely 2) at the smaller campus. Saturday and holiday staffing is provided by 1 nurse who covers both hospitals. An additional team member may be called in if needed, as determined by the VAT members. During the time of implementation reviewed in this quality improvement initiative, the team included 4 experienced vascular access nurses, 1 vascular access nurse with outside experience but newer to the team, and 1 nurse with bedside experience at the hospital but new to inserting PICCs. The team maintain a list of patients that are no longer able to have bedside PICC insertion due to considerations such as stenosis, deep vein thrombosis (DVT), dialysis etc. Interventional Radiology is consulted for these cases and placed an average of 19 PICCs per month in the Cardiac Catheterization Lab under fluoroscopy. Care and maintenance of all catheters is performed by bedside staff in the organization, with the VAT only involved with insertion and initial assessment and some problem solving for device performance issues when consulted.

Device performance beyond CLABSI is measured on an on-going monthly basis by the VAT. Aggregation of outcomes data was standardized in 2017. Members of the team are accountable for reviewing the medical records of the patients in whom they have placed vascular access devices to determine disposition of the catheter at the time of discharge and to capture inpatient line days (for PICCs, midline catheters and peripheral intravenous [IV] lines placed by the team). This quality improvement project is within the standard scope of the Infection Prevention Program following evidence-based recommendations and therefore approved by the Infection Control Committee, Shared Governance and Medical Council. It did not require institutional review board approval.

Methods

CLABSI surveillance is performed by a certified (CIC) member of the infection control team, strictly following the current Centers for Disease Control and Prevention (CDC) National Healthcare Safety Network (NHSN) Protocol for Central Line and Non-Central Line Associated Bloodstream Infections.⁹ All positive cultures are screened to determine if they have been obtained from an inpatient or a patient who has been recently discharged and then evaluated against the surveillance definition. After determining that a bloodstream infection meets the criteria for a primary, laboratory-confirmed bloodstream infection, vascular access device use is reviewed to determine whether the infection meets criteria for a CLABSI, which requires that a central line has been present for greater than 2 days before infection onset and in situ the day of or day before the infection. Any questions on CLABSI surveillance are directed in writing to the CDC for clarification; there is no institutional adjudication outside of the Infection Control Department within the organization, consistent with surveillance expectations.

The VAT maintains prospective records on each PICC that they insert. This database serves as the denominator source for total number of PICCs placed by the team as well as PICC days. The team reviews the electronic medical record to verify device removal and final disposition of the catheter and whether it remained in place at discharge. Device days are counted following the CDC convention, with insertion day being counted as day 1. Complications are assessed primarily via chart review after patient discharge, with information on reason for device removal obtained through the documentation charted by the primary nurse or VAT consult. The complication rates represented are only those that resulted in device removal.

CLABSI rates are calculated monthly by the infection control team and distributed throughout the organization. Device performance is aggregated monthly by the VAT mainly for their internal reference but included in the Infection Control Annual Report as well. These combined measures were continuously reviewed during the postimplementation period and aggregated for statistical purposes at regular intervals to establish risk reduction and harm avoidance.

Statistical analysis was performed using the analysis features within the CDC's NHSN statistics calculator, which includes tools to compare 2 incidence density rates and generate a P value. Significant results were considered at P<0.05. Complication rates are expressed in frequency and percentages. Percent reduction is calculated using Microsoft Excel by expressing the difference in values over the initial value to generate the percentage change.

Results

Baseline bloodstream infection rates for PICCs placed by the team from January to September 2017 was 1.83/1000 PICC days (11 occurrences, in 6012 measured days; Table 2). Statistically significant CLABSI reduction in the PICCs placed by the team was achieved within 6 months of implementation (0 infections in 2165 PICC days; P=0.0339). The first full year of implementation demonstrated an 80% reduction (P=0.02) from baseline in PICC CLABSIs, with a yearend rate of 0.374/1000 PICC days. Data are presented on 30 months of postimplementation findings, totaling 1375 devices and 12,352 inpatient central line days. With the inclusion of this entire period, postimplementation the CLABSI rate was calculated at 0.162/1000 PICC days, a 91.15% reduction from baseline(P=0.0002). The second full calendar year of implementation had 0 reported CLABSI.

Overall completion of therapy for the PICCs over the 30-month period was 94.4%. This number encompasses catheters removed due to completion of therapy, patient discharge (with or without the PICC remaining in place), and patient death. The highest complication rate was due to device dislodgment at 3.93%. No other complication exceeded 1% as a reason for removal. DVT rate reflects only symptomatic DVTs as recorded in the medical record. Details are included in Table 3. Lumen distribution during the same period was 41.62% single-lumen, 32.42% double-lumen, and 25.95% triple-lumen PICCs.

Discussion

Despite adherence to maximum sterile barrier precautions, the baseline infection data observed at the organization were more than historic rates noted in the literature of 1.0/1000 central line days for PICCs.¹⁰ The majority of infections observed occurred at least 1 week after insertion, which suggested elements of care and maintenance rather than insertion considerations were likely contributing to the incidence of infection during individual case review. The absence of available single-lumen devices may have also increased the risk of infection, which was addressed and corrected earlier in the year (during baseline measurement), before implementation of the new device.^11,12 Because the organization relies heavily on the VAT to place inpatient devices (at the time of implementation, the team was placing primarily PICCs and mid-lines, but over the past 2 years have also been consulted more frequently for difficult access peripheral IVs), it was anticipated that a reduction in PICC CLABSIs would positively impact the organization's overall CLABSI performance. Since the full implementation, the standardized infection ratio for the organization has remained below 1.0 (which is the reference number indicating the performance is as predicted based on case mix, bed size, central line use, and teaching affiliation). In 2018 (the first full calendar year of use), there were a total of 4 CLABSIs involving PICC lines. One of them was a device present on admission, so it is excluded from the calculations below (numerator as well as denominators), as it was not one of the catheters placed by the VAT. Two infections occurred inpatients in whom it was the only present vascular access device, and all other sources of infection were ruled out, so they are included in the counts below. One patient had PICC as well as a dialysis catheter. Using the basic methodology that the CDC applies to other infections (specifically surgical site infections), this infection is not included in the PICC CLABSI count, as dialysis catheters have a much higher associated rate of infection. Plus, there was documentation in the record of some catheter-related complications with that dialysis catheter at the time of the bloodstream infection.10 (Please note, both of these “excluded” infections are still counted and included in public reporting by the organization; they are just not cases for the purpose of calculations on the impact of device implementation.)

The main objective of this product introduction was to address the incidence of CLABSI in PICCs placed by the VAT (excluding those placed by Interventional Radiology or the Neonatal Intensive Care Unit). Statistical significance in this CLABSI reduction was achieved within 6 months, and each calendar year after implementation, the rate dropped lower, ultimately achieving an aggregate reduction of 91.15% incidence density rate. Had the infection rate continued as the baseline performance, the organization would have been predicted to experience 23 PICC CLABSIs compared with the 2 that were identified. Using a conservative mortality estimate of 20%, this evidence-based practice change may have contributed to avoiding 21 CLABSIs and 5 patient deaths.

Robust implementation of a midline catheter program was completed within the organization in 2016, before the implementation of the (AM/AT) PICCs, with continued shifts toward peripheral vascular access options when appropriate and proportional decreases in the numbers of PICCs inserted. Policies and order sets were updated to guide device ordering practices earlier in the year of implementation of the new catheter. PICC-associated CLABSI rates are expressed per 1000/PICC days per standard convention to reflect incidence density based on exposure.

There was 1 other focused vascular access improvement effort which was occurring at the same time as the implementation of the (AM/AT) catheter which was an emphasis on improved vascular access dressing integrity, including the addition of gum mastic liquid adhesive on dressing margins. During the postimplementation period, a convenience sample of 1868 observations on PICCs were obtained through enhanced quality audits and noted to have fully intact dressings 97.27% of the time. It is possible that the improved dressing adherence (baseline data was available for peripheral IVs and was only 55% fully intact), and additional clinical rounding on all vascular access devices may also have contributed to the observed improvements; however, the same immediate, temporal improvements in bloodstream infection were not observed in the small numbers of non-AM PICCs that continued to be placed nor the other central catheters and peripheral catheters in use. Because the near elimination of infections was noted only in these specific PICCs, it is less likely that the dressing adherence and increased rounding were the significant contributors to the observed improvements, although they may have played some role.

Limitations

This was a single-center study in a large community hospital. Inserters were largely experienced vascular access nurses, but care and maintenance remain the responsibility of the bedside nurse. Complications are primarily obtained through review of the electronic medical record and only reflect those events that are noted to have resulted in device removal, so may underrepresent the true incidence of some occurrences, particularly occlusion or asymptomatic DVT, without further detail to lend clearer understanding for unlikely outcomes (such as infiltration without dislodgement) and are without consistent definition for the bedside clinician. The first year that vascular access device complications beyond CLABSI were collected and analyzed systematically was 2017, making a larger baseline period impractical. Additionally, complication stratification by number of lumens was not routinely undertaken, which would have provided further insight for additional opportunities for improvement.

Conclusions

Combined with continued compliance with basic prevention strategies (ie use of a central line insertion checklist/insertion bundle) and optimization of device selection and lumen justification, the introduction of an (AM/AT) PICC was associated with a significant reduction in CLABSI.