In the United States more than 5 million patients each year get a central venous catheter. Indications for central venous catheterization include infusion of irritant drugs like chemotherapy or total parenteral nutrition, poor peripheral venous access and long term administration of drugs such as antibiotics. This ubiquitous procedure has many associated complications that result in morbidity, mortality, and increased healthcare cost. The overall complication rate is more than 15% and a great preventive effort is done.
Catheter Related Bloodstream Infection (CRBSI) is a serious and feared complication associated with prolonged hospital stays, increased costs and risk of mortality. CRBSI is defined as the presence of bacteremia originating from an intravenous catheter. A lot of effort is done to reduce CRBSI including heightened attention to hygiene in placement and care, improved education and training, and placement of a team with specialized skills. Adherence to best practice for central line placement is shown to reduce risk of CRBSI .
Moreover, central venous catheters (CVCs) are associated with deep vein thrombosis (DVT) and pulmonary embolism. Besides interruption in treatment, catheter-related DVT increases morbidity and mortality. Cancer and admission to intensive care are independent risk factors. Existing data report wide estimates of this adverse outcome, ranging from less than 1% to as high as 38.5%, dependent on the population studied, method of diagnosis, and use of prophylaxis measures.
Peripherally inserted central catheter (PICC) is a CVC which placement and use have been widespread since it was first described in 1975. It is a well-established alternative to CVCs placed via subclavian or jugular veins. It is easy to place, safe and cost-effective compared to others often used central lines. PICC is inserted via a peripheral vein in the upper arm and terminates like other central lines in the vena cava superior. Placement and use is associated with few complications .
Another peripherally inserted catheter is the midline which by definition is 7.5 to 20 centimeters long (3-8 inches) and thus not a CVC. The midline catheter was introduced in 1950s. It has undergone major improvement in material technology and techniques for achieving vascular access. It is inserted in the same peripherally veins as the PICC, but the tip is advanced no further than the distal axillary vein and is therefore classified as a peripheral intravenous catheter with corresponding advantages and disadvantages. The midline cannot be used to vesicants or irritants like most chemotherapy, vasoactive agents, or medications with extremely low or high pH values. The midline is suitable for use from 5 days until 4 weeks to drugs and solutions, which safely can be administrated in a peripheral venous catheter. Severe complication to placement and use of midline is rare, but due to previous problems primarily related to the midline catheter material, its use is limited .
In a large review from 2006 the incidence of CRBSI among in- and outpatients having PICC or midline were estimated to 3.1% (95% confidence interval (CI) 2.6-3.7) and 0.4% (95% CI 0.0-0.9), respectively. It has been demonstrated that CVCs can be decreased through the use of midline catheters . A retrospective descriptive review from two hospitals in America showed the effectiveness of implementing a midline program resulted in a 78% reduction in CVC line-associated bloodstream infection. In a similar Australian retrospective cohort study in a ventilator unit population, a significant decrease in the rate of CVC line-associated bloodstream infections was found after use of midlines. It seems that the introduction and regular use of midlines when warranted may reduce the overall incidence of CRBSI and its sequelae in certain hospital environments. In a meta-analysis including 11,476 hospital admitted patients with PICC, DVT was found in 3.44% (95% CI 2.46-4.43). DVT occurrence in relation to midline is understudied, but is reported with a low incidence between 0-2 %. The overall incidence of DVT and related potentially secondary complications of both catheters seems low. The risk of minor complications such as pain, leakage or phlebitis is in a retrospective comparison study found to be 11.5% for midlines and 1.5% for PICC, respectively (P\<0.001).
The efficacy of the PICC is well studied, the incidence of side effects is known and its use is implemented all over the world. However, the efficacy of midlines compared to PICCs has not been evaluated prospectively. The present study aim to examine the efficacy and safety of midline catheters using standard care with PICC as reference.
Patients eligible for screening for inclusion are identified among all patients where staff from a general ward requests a central line. The randomization is 1:1 between the PICC (control group) and a midline catheter (intervention group). After placement the patients will be closely followed until the day of removal of the catheter. To obtain information on length of hospital stay and mortality, the electronical journal will be checked until 90 days after catheter removal. The incidence of complications that occur will be registered and the two catheter groups will be compared.
Risk and benefits by participating in the trial: In patients with poor peripheral venous access or need of long time administration of medicine or fluids a central line is routinely placed. The puncture site on the upper arm and placement technique are the same for both catheter types. Therefore the complications during placement are expected to be independent of catheter type. PICC require x-ray confirmation of tip placement, leading to additional costs and exposing the patient to unnecessary radiation. X-ray verification of tip position is not necessary when placing the midline catheter. By participating in this study the patients have benefit of more close observation and patients in the intervention group have the possible benefit of reduced risk of having a CRBSI. In the intervention group, the risk of minor complications like pain in relation to fluid or medicine administration, infiltration or phlebitis is expected to be increased compared with the control group. The discomfort in relation to the minor complications is expected to disappear without any persistent complications.
Design: Single-center randomised controlled trial with in- and out-patients from medical and surgical departments.
Catheter placement: The informed consent and randomization is performed prior to placement of catheters. With the patient supine ultrasound is used to identify the desired vein on the relevant upper extremity. Full sterile technique is used and includes the operator wearing sterile gown, mask, cap, and sterile gloves. The area is then prepared with chlorhexidine followed by adequate sterile draping. The Seldinger technique is used to insert the catheter. Successful placement in a vein is secured by aspirating blood from the catheter. The catheter is then flushed with minimum 20 mL saline. Depending on the randomization a PICC is placed and the tip position in vena cava superior is verified by a chest x-ray or a midline is placed without the need of x-ray verification. The patient is then returned to the medical or surgical ward.
Recruitment procedure: Subjects are recruited among patients where staff from a general ward requests a central line. The anesthesiologist or special trained nurse responsible for inclusion will apply to the national regulations regarding informed consent to participation in a clinical trial. Hence, in addition to oral information the potential participant will receive written information including both the specific information on the current study as well as the general information pamphlet on participant rights when entering a clinical trial. All information and inclusion will be given by physicians or special trained nurses who possess the sufficient professional prerequisites to be authorized by the sponsor to have a direct involvement in the project. Information will be given in private and the participant will be allowed to have an assessor present. The participant will be given a brief reflection period before making their decision. As always it is voluntary to participate and the subjects can withdraw their commitment to participate at any time.
Baseline variables: Date of birth, age, gender, ethnicity, height, weight, and medical history.
Procedure related variables: Date of randomization, date placement, time used for placement, number of skin punctures, type and length of placed catheter, name of the access-vein, accidental arterial puncture, bleeding complications, and tip placement on chest x-ray (control group only - assessed from an anterior and posterior X-ray of thorax).
Registrations at the wards: CRBSI, deep vein thrombosis, catheter removal date and cause.
Phlebitis scale score: 0 No symptoms; 1 Erythema at access site with or without pain; 2 Pain at access site with erythema or edema; 3 Pain at access site with erythema or edema; streak formation; palpable venous cord; and 4 Pain at access site with erythema or edema; streak formation; palpable venous cord \< 2.5 cm Infiltration scale score: 0 No symptoms; 1 Skin blanched; edema \< 2.5 cm in any direction; cool to touch; with or without pain; 2 Skin blanched; edema in 2.5 to 15 cm in any directions; cool to touch; with or without pain; 3 Skin blanched, translucent; gross edema \> 15 cm in any directions; cool to touch; mil-to-moderate pain; possible numbness'; and 4 Skin blanched, translucent; skin tight; leaking: skin discoloured: bruised; swollen; gross edema \> 15 cm in any direction; deep pitting tissue edema; circulatory impairment: moderate-to-severe pain; infiltration of any amount of blood product, irritant, or vesicant.
Sample size estimation and power calculation: The primary outcome is CRBSI. The power calculation is based on an expected incidence of 5% in the PICC group with reference to the literature and an expected incidence of 0% in the midline group with reference to a follow-up of the first 107 midline catheters inserted in patients at Aalborg University Hospital from the 5th of October 2017 to the 26th of February 2018. With an alpha of 0.05 and a beta of 0.2 (power 0.8) the sample size is 304 with 152 patients in each group.
Statistical methods: Descriptive data will be presented in a baseline table according to catheter type. For normally distributed measurements the differences between groups will be compared using Student´s t-test. Variables considered not to be normally distributed will be analysed by Mann-Whitney´s U-test. The results from primary and secondary endpoints will be presented in a separate table also according to catheter type. The differences between groups will be compared using Wilcoxon two-samples test/Fisher´s exact or unpaired t-test. Statistical analyses will be performed using STATA software (version 14; STATA, Corporation, College Station, TX). A two-sided P value of less than 0.05 will be considered statistical significant.
