A comparison of infections and complications in central venous catheters in adults with solid tumours

Introduction

Establishing and maintaining venous access is an important part of managing oncology patients (1) given that treatments are increasingly complex and prolonged. Combining chemotherapy with biological or targeted treatments and other supportive intravenous products necessitates long-term venous access devices. These devices minimize the discomfort of frequent cannulation while preserving the peripheral veins. Central venous catheters (CVCs) include totally implantable (central venous ports or ports) tunnelled catheters and peripherally inserted central catheters (PICCs) (2).

Central venous access is required for administration of intravenous chemotherapy via ambulatory pumps, and for frequent or vesicant chemotherapy. Schiffer and colleagues found that there is insufficient evidence to recommend a specific type of CVC routinely for patients with cancer (3). Therefore, it is important that oncologists discuss CVC options, including risks and benefits with the patient. The aim of this study was to compare infection and complication rates among the three most common CVCs used in this centre to guide future clinical practice.

Methods

This study was conducted on a chemotherapy unit in Leicestershire commencing 1 January 2011 and ending 1 August 2013, for the duration of 18 months in total. This unit administers approximately 1800 doses of chemotherapy and anti-cancer drugs per year, the majority of which are administered intravenously. All adult oncology patients with solid tumours receiving chemotherapy who required central line insertion were included in this study (Tab. I). Haematology patients with CVCs were excluded. Data including patient demographics, nature of their tumour, chemotherapy regime, CVC type and duration in situ, infection rates and complications, and patient mortality were recorded prospectively. Where patients suffered complications, infections, or death, further details were obtained accordingly. CVCs were divided into two primary groups; ports and non-ports; the latter group was then subdivided into tunnelled catheters and PICCs. The primary outcomes measured were the rate and type of complication, while the secondary outcome was the complication-free CVC duration. Data were compared between the different types of CVCs used in the study. Multiple entries were recorded if the patient had more than one line during their treatment. Patients were followed-up until the device was removed, mortality occurred, or the end of the study.

CVC infection was defined as any infection related to a vascular catheter including local (e.g. insertion or exit site) and systemic (e.g. bloodstream) infections with positive cultures (4). Our hospital uses the criteria by the Hospital Infection Control Practices Advisory Committee (HIPAC): A catheter-related blood stream infection (CLABSI) is a primary blood stream infection (BSI) in a patient who had a central line within 48-h period before the development of the BSI and is not bloodstream related to an infection at another site (5). Catheter displacement was defined as the displacement or migration of the CVC line from its original site. Complete occlusion was defined as when it became impossible to infuse intravenous fluids into the catheter, or to withdraw blood from it.

All vascular access devices were inserted using local anaesthetic with ultrasound and fluoroscopic guidance. Lines were implanted in the angiography theatre to ensure the highest standards of asepsis. PICC and tunnelled catheters were inserted by two consultant radiologists. The ports were inserted by a single consultant radiologist. PICC lines were inserted into the brachial veins, tunnelled catheters into the subclavian or internal jugular veins and central venous ports exclusively into the right internal jugular vein with the port sited over the right second rib. PICC lines used were 4Fr polyurethane single-lumen catheters inserted with a micro seldinger technique and secured with an adhesive statlock dressing. Tunnelled cuffed silicone catheters were 6Fr single-lumen nonvalved catheters and secured with 3.0 prolene sutures left in situ for at least 1 month to allow time for tissue ingrowth. The cuff was placed 1-2 cm proximal to the skin exit site, at least 2 cm deep and always placed at a minimum distance of 30 cm from stomas. The PICC lines and tunnelled cuffed silicone catheters were inserted by two consultant radiologists. The central venous ports were connected to 6Fr single-lumen nonvalved silicone catheters. Type of line inserted was influenced by patient choice, chemotherapy regime, expected duration of treatment and suitability of anatomical factors such as body habitus that may favour one device over another. For short-term chemotherapy regimens (between 3 and 6 months), PICC lines were predominately inserted. For medium to long-term regimens over 6 months, tunnelled lines or ports were inserted as per current guidelines (6).

The routine practice at this institution is to flush tunnelled cuffed catheters and PICC lines weekly when not in use, using 5 ml heparinised saline (Hepsal 10 I.U/ml) 10 ml syringe, needle-less valves, alcoholic Chlorhexidine 2% (Clinell) cleanser to decontaminate catheter hubs, and aseptic technique by hospital-trained nurses. Twenty-one out of the 28 patients tunnelled cuffed catheters and PICC lines were flushed by community nurses/patients on non-treatment weeks. Written patient advice sheets and additional verbal advice is given to all patients on caring for their external lines. Hibiscrub plus is supplied for washing around the exit sites, showering is encouraged rather than bathing and swimming is not allowed. Dressings covering tunnelled exit sites are changed daily after shower. PICC line outer dressings are changed weekly or if soiled or wet; PICC dressings used to secure the line are changed as required. Strict hand hygiene prior to accessing all CVCs is reinforced. All ports are flushed monthly when not in use with heparinised saline (Hepsal 10 I.U/ml) using aseptic technique by chemotherapy nurses, following national Epic2 guidelines (4).

Statistical analysis

Data were compared using the Chi-squared test and independent sample t-test, with a p value of less than 0.05 indicating a statistical significance. Median age of patients and duration of insertion of catheters were recorded in years and days, respectively, while the complication rate was recorded as the number of complications per 1000 catheter days.

Results

During the study period, 19 tunnelled catheters were inserted in 17 patients, nine PICC lines in eight patients and 30 ports inserted in 30 patients, totalling 58 central lines in 55 patients. The median age for all CVCs was 54 years (range 37-80 years). The number of ports compared equally between both genders; 10 males and nine females had tunnelled catheters and eight females had nine PICC lines inserted. Chemotherapy regimens were comparable between all the three groups. However, flushing of lines was four times more frequent in tunnelled catheters and PICC lines than in central venous ports, as per local protocols.

In the central venous port group of 30 patients (30 lines), the median age was 59 years, with a median duration of insertion of 149 days. None of the lines had a documented infection, although one port eroded through the skin, most likely because the patient had limited subcutaneous fat.

In the noncentral venous ports group, the overall median age was 51.1 years, and the median line of duration was 77 days. There were seven infections and five line displacements in total. Due to a higher complication rate in this group, we subdivided the group results according to whether the patient had a tunnelled catheter or a PICC line inserted.

The PICC line subgroup had eight patients (nine lines), with a median age of 50.7 years and a median duration of 52 days. Three had complications, two displaced (after 5 weeks and 12 weeks) and one became permanently occluded; there were no documented infections.

The tunnelled catheter subgroup included 17 patients (19 lines) with a median age of 51.3 years and a median duration of insertion of 89 days. Ten lines had complications, three displaced (after 4 weeks and 1 week) and seven were removed due to infection.

Three of the seven patients with infected tunnelled catheters developed systemic sepsis requiring hospital admission for intravenous antibiotics. Two of these patients developed positive blood cultures and five had positive swab cultures around their line exit sites. Organisms isolated in the blood cultures were Pseudomonas species, Klebsiella pneumonia and Escherichia-coli. Organisms isolated in exit sites were Coagulase negative Staphylococcus in four patients, and Pseudomonas species in three occasions. Two of the tunnelled catheters became infected within 28 days of line insertion. Six of the seven patients who developed infected tunnelled catheters were colorectal cancer patients (five of whom had stomas) and the remaining patient had a soft tissue sarcoma (Tab. II).

Discussion

Central venous access is an essential part of the management of oncology patients, but the presence of an intravascular foreign body connected to the outside environment exposes patients to the risk of life-threatening infections. Every year, almost 6000 patients in the UK acquire a CLABSI. Oncology patients are further susceptible to infection because of their compromised immune system and every attempt should be made to minimize their risk of developing sepsis (1). A CLABSI is a serious infection that occurs when bacteria enters the bloodstream through the central line. CLABI results in thousands of deaths each year and a great cost to the health service (5). Risk factors for CLABSIs can be intrinsic, nonmodifiable characteristics such as patients age, underlying diseases or conditions; or extrinsic, modifiable factors associated with CVC insertion, type of CVC or maintenance. The extrinsic risk factors have the greatest impact on risk of CLABIs (7).

Randomized controlled trials (RCTs) report that CLABSIs are largely preventable and individualized patient and career education decreases infection rates. Therefore, the use of a CVC care bundle is recommended (3, 5). Within this study, of the six patients who had infected tunnelled catheters, four had stomas and all six flushed their lines themselves at home or had district nurses on non-treatment weeks. Tunnelled catheters were all inserted at the recommended distance (30 cm) from the stoma site. High infection rates in this group are therefore likely secondary to contamination through handling of the stoma. Patients should be given comprehensive verbal and written instructions on flushing and care of their central lines especially in the presence of a stoma.

A previous study on paediatric patients with leukaemia that compared tunnelled catheters with central venous ports found that CVCs have a three to seven times higher rate of CLABSIs than ports (8). A similar picture is presented in adults with solid tumours undergoing infusional chemotherapy with a 4.9 times higher rate of complications with tunnelled catheters than ports, with infections the most common complication (9). A further study found similar findings for oncology patients with a median of 0.2 infections per 1000 catheter days for oncology patients with ports versus 1.4-2.2 infections per catheter days for tunnelled catheters (3). In addition, HIPAC examined multiple studies concluding that totally implanted devices had the lowest reported infection rates compared with tunnelled and non-tunnelled CVC (4). However, there is a paucity of documented studies regarding the best method of administration for long-term chemotherapy.

Patients are given options on their type of CVC; chemotherapy nurse specialists (CNS) discuss the differences in CVCs and provide literature and guidance. Benefits of ports are that patients can swim and bath as normal. Minimal maintenance of lines is required: no change of dressing around exit sites, monthly flushing to maintain patency when not in use, compared with weekly flushing for tunnelled catheters and PICC lines. There is no visibility of an external line and a minimal risk of damage/pulling. The acceptability to patients is greater with a port. As one patient said, ‘It allows me to feel normal in between treatments’. The option of an implanted catheter promotes shared decision making between patient and healthcare professionals at the beginning of treatment.

Following this study, we have introduced improved patient education concerning CVCs. Further, hand decontamination education will now be given to all patients choosing to flush their own CVCs at home, particularly stressing the importance of cross contamination when patients have stomas. Current national and international guidance consistently identifies that effective hand decontamination results in significant reductions in the carriage of potential pathogens and decreases the incidence of preventable hospital-acquired infection (HAI) leading to a reduction in patient morbidity and mortality (4).

We recommend further clinical studies on infection rates on CVCs documenting the effects of improved, individualized education for patients self-flushing for tunnelled catheters and PICC lines compared with ports. Effective research-based education should remain a priority in reducing CLABSIs. One limitation of this study includes the use of CLABSI to define infection. Catheter-related sepsis (CRBSI) is a clinical definition sometimes used interchangeably with CLABSI; however, a diagnosis of CRBSI requires specific laboratory testing, for example catheter tip cultures that accurately identify the catheter as the source of the BSI (6). Another potential limitation of this study is that tunnelled lines were secured using a suture rather than a sutureless device, as level II evidence exists that this may increase the risk of infection (6). Both retrospective and prospective data were recorded; however, an RCT comparing outcomes in different CVC lines would be preferable. A relatively small number of patients were monitored, and this could be improved by concurrently monitoring CVC complication rates in another chemotherapy unit. Cost analysis was difficult to conclude accurately, therefore it was not undertaken. A further study, evaluating the effect of education on caring for CVCs, comparing infection rates would be recommended.

Conclusion

We have demonstrated that in this single centre, tunnelled catheters and PICC lines when compared with ports have a significantly higher complication and infection rate. Implantable ports appear to be a safe, aesthetically acceptable, venous access device for mid to long-term infusional therapies in oncology patients. This, in line with previous evidence, should guide future clinical practice, giving patients a further option to external CVCs.

TABLE I -

Demographics of the different types of cancers included in the study

Cancer type	Port-a-cath	Non port-a-cath	p
GBM = glioblastoma multiform.
Colorectal	17	15	0.002
Breast	10	3	0.48
Ovarian	1	4	0.54
Anus	0	1	0.63
0esophagus	0	1	0.63
GBM	0	1	0.63
Sarcoma	2	1	0.31
Unknown primary	0	2	0.58
Total	30	28	0.03

TABLE II -

Demographics of patients included in the study

	Port-a-cath	Non port-a-cath	p
Total number of CVC	30	28
Gender (Male)	15	10
Median age (years)	56.7	51.1
Duration in situ	154	77.5	0.006
Total number of complications	1	13	0.74
CVC infection	0	7	0.62
Displacement of line	0	5	0.019
Blocked line	0	1	0.313
Patients with stomas	11	10	0.03
CVC infected patient with stoma	0	5
Hospital care of CVC	30	7	0.001
Patient admitted for IV antibiotics	0	3
Colorectal patient with CVC infection	0	6
Sarcoma patient with CVC infection	0	1

• 1. Galloway M. Insertion and placement of central catheters in the oncology patient. Semin Oncol Nurs 2010; 26: 102-112
• 2. Lyon SM.,Given M.,Marshall NL. Interventional radiology in the provision and maintenance of long-term central venous access. J Med Imaging Radiat Oncol 2008; 52: 10-17
• 3. Schiffer CA.,Mangu PB.,Wade JC. Central venous catheter care for the patient with cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol 2013; 31: 1357-1370
• 4. Pratt RJ.,Pellowe CM.,Wilson JA. epic2: National evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J Hosp Infect 2007; 65: Suppl 1-S64
• 5. O’Grady NP.,Alexander M.,Burns LA. Healthcare Infection Control Practices Advisory Committee. Am J Infect Control 2011; 39: S1-S34
• 6. 2011 guidelines for the prevention of intravascular catheter-related infections.
• 7. Lu SF.,Chen JH.,Shang WM.,Chou SS. [Prevention and nursing care of central line-associated bloodstream infections in critically ill patients]. Hu Li Za Zhi 2012; 59: 5-11
• 8. Alexander N. Question 3. Arch Dis Child 2010; 95: 239-241
• 9. Ng F.,Mastoroudes H.,Paul E. A comparison of Hickman line- and Port-a-caths- associated complications in patients with solid tumours undergoing chemotherapy. Clin Oncol (R Coll Radiol) 2007; 19: 551-556