Peripherally inserted central catheters (PICCs) are often used for patients requiring short-term venous access (eg, ≤5 days), including intravenous antibiotics. Reference Paje, Conlon and Kaatz1 However, PICCs are associated with risk of deep vein thrombosis (DVT) Reference Chopra, Anand and Hickner2 and bloodstream infection. Reference Chopra, O’Horo, Rogers, Maki and Safdar3 Midline catheters, which appear to have a lower complication rate, Reference Xu, Kingsley and DiNucci4,Reference Mushtaq, Navalkele and Kaur5 may be an option for some patients. However, midline catheters have evolved, with newer materials and design, Reference Cawcutt, Hankins, Micheels and Rupp6 but data regarding indications for use, patient experiences, and adverse events remain limited. Reference Xu, Kingsley and DiNucci4,Reference Chopra, Kaatz and Swaminathan7 To bridge this gap, we compared indications for use as well as patient-reported and chart-documented complications for a cohort of patients that received midline catheters and PICCs.
Our primary objectives were to assess: (1) indications for device placement, (2) percentage of patients reporting a potential device-related complication, and (3) complications documented in the electronic medical record (EMR) during the same time frame.
Methods
As part of a study examining patient-reported experiences with PICCs, Reference Krein, Saint and Trautner8 we performed a prospective observational study comparing indications for use and complications among patients receiving a midline catheter or a PICC from August 2015 through May 2017 at an urban safety net hospital (ie, a hospital providing a significant level of care to patients regardless of their ability to pay). A convenience sample of hospitalized patients was used. Patients were eligible to participate if they: (1) had a new midline catheter or PICC placed within 3 days of enrollment, (2) were ≥18 years old, and (3) were able to speak English or Spanish. Patients were excluded if they were unable or refused to provide consent or had previously participated in this study.
On average, during the study period, 111 midline catheters and 120 PICCs were placed monthly by the hospital vascular access nurse team, who used the Michigan Appropriateness Guide for Intravenous Catheters (MAGIC) Reference Chopra, Saint and Woller9 criteria for device selection. Midline catheters were 10-cm Bard Powerglide 18-gauge catheters (Becton Dickinson, Franklin Lakes, NJ) inserted under ultrasound guidance. Double-lumen Bard PowerPICC catheters (Becton Dickinson) were used in the inpatient setting, and single-lumen catheters were used for home infusion.
The study was approved by the health system institutional review board (protocol H-36119).
Data collection
Data regarding indications and complications were collected from patients and via a review of the electronic medical record (EMR). Interviews with patients were conducted at enrollment and at 14, 30, and 70 days after device placement. During follow-up assessments, patients were asked structured questions to determine whether the device was in place, whether another device had been inserted, and whether they had signs or symptoms of a complication potentially related to the device. They were asked to reflect on the prior 7 days at the interview on day 14 and on the prior 30 days at the interviews on days 30 and 70. Patients were also asked to share any other problems with the device. Study staff reviewed the EMR during the same 70-day time frame and collected information on insertion and removal dates, number of devices placed, and complications. Documentation of a DVT or bloodstream infection required an explicit statement of the condition by a medical provider in the EMR.
Data analysis
We conducted a descriptive analysis. Characteristics of patients receiving midline catheters versus PICCs were compared using the Fisher exact test. All statistical tests were 2-sided and a P-value of 0.05 was considered significant. All analyses were performed using Stata MP version 15.1 software (StataCorp, College Station, TX).
Results
Of the 68 patients eligible after midline catheter placement, 58 consented (85.3%). Of those, 50 were included in the analysis. We excluded patients with no response to the initial interview or to any of the 3 follow-up interviews. In total, 63 patients with PICCs hospitalized on the same inpatient units during the same period were selected as a comparison group. We did not detect any statistically significant differences between groups in terms of age or race (Table 1). We identified a significant difference, however, with respect to sex; the midline catheter group had more women than the PICC group. The most commonly reported reason for midline catheter insertion was difficult venous access (52.0%), and chemotherapy was the most common reason for PICC insertion (65.1%). Although not statistically significant, 20.0% of patients with midline catheters reported experiencing pain, discomfort, bleeding or other trauma during insertion compared with 31.8% of those with PICCs (P = .144). The device dwell time was ≤5 days for 50.0% of midline catheter patients and for 46.0% of patients with PICCs. Of those with the device for ≤5 days, difficult venous access was the indication reported for device insertion in 56.0% of patients with midline catheters versus 13.8% of those with PICCs (P = .001).
Table 1. Baseline Patient and Device Characteristics
Note. PICC, peripherally inserted central catheter.
a Fisher exact test.
b Information derived primarily from chart review data.
Complications based on patient report and medical records are listed in Table 2. One midline catheter patient reported seeing a doctor for signs suggestive of an infection and was told that they had a bloodstream infection due to the catheter, whereas 6 PICC patients (9.5%) reported signs of potential infection requiring them to see a doctor, but none reported being told they had a bloodstream infection. Compared to those with midline catheters, more patients with PICCs reported minor complications, such as redness at insertion site or removal difficulty. No patients with midline catheters were documented as having a DVT in the chart, compared to 14.5% of patients with PICCs. Likewise, none of the midline catheter patients had a bloodstream infection documented in the chart, whereas 1 patient with a PICC did.
Table 2. Complications up to 70 Days After Initial Midline or PICC Placement
Note. PICC, peripherally inserted central catheter.
Discussion
We report 2 main findings of our study. First, in our study population, difficult venous access was a primary indication for patients with midline catheters, not PICCs. Second, patients with midline catheters reported fewer potential complications compared to those with PICCs. Likewise, we found no EMR documentation of serious complications among midline catheter patients, supporting prior studies reporting that complication rates are lower with midline catheters than with PICCs. Reference Xu, Kingsley and DiNucci4,Reference Mushtaq, Navalkele and Kaur5
Our findings add to a growing evidence base suggesting that midline catheters may be a viable and safer alternative to PICCs for patients who require short-term venous access for peripherally compatible therapies. MAGIC Reference Chopra, Saint and Woller9 recommends a midline catheter over a PICC if the proposed duration of a peripherally compatible therapy is ≤14 days. Accordingly, at our study site, the vascular access nurses call the ordering provider when a PICC request does not meet MAGIC criteria and, if appropriate, they recommend a midline catheter. This approach, our findings suggest, has led to more appropriate (and possibly safer) device use with 52.0% of patients having a midline catheter for an indication of difficult venous access versus 11.1% of patients with a PICC. Other sites implementing midline catheter programs targeting patients with difficult venous access have also achieved lower rates of PICC placement. Reference DeVries, Lee and Hoffman10
Our study has several limitations. First, this was a small study of patients recruited from a single hospital, so results may not be generalizable to other patient populations. Also the midline catheter group had more women than the PICC group, which may have affected the results. Sampling was not random, and data collected by interviewing the patient has the potential for recall bias. Also, we only reviewed medical records at the study hospital and affiliated clinics, so data on documented complications may be incomplete if patients received care elsewhere. In addition, EMR-derived complications were based on provider documentation and not objectively verified.
Limitations notwithstanding, our study revealed that the primary reason for midline catheter insertion was difficult venous access, whereas chemotherapy infusion was the most common reason provided for requiring a PICC. Midline catheters also appear to be potentially effective options for short-term venous access.
Acknowledgments
We thank the patients who so graciously participated and provided data for this study. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Department of Veterans’ Affairs or the US federal government.
Financial support
This material is based upon work supported by the Department of Veterans’ Affairs, Veterans’ Health Administration, Office of Research and Development, Health Services Research & Development Service (grant no. IIR 12-395 and RCS 11-222). Dr Trautner’s work is supported in part by the Houston Veterans’ Affairs Health Services Research & Development Center for Innovations in Quality, Effectiveness and Safety (grant no. CIN 13-413).
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
