Venous cannulation is an essential requirement in intensive care unit (ICU) patients and is most commonly performed in an antegrade direction with the direction of the blood flow. Stasis of blood around the catheter, particularly in the angle between the vein wall and the catheter, in addition to catheter-induced phlebitis, may initiate a thrombosis. Thrombosis can extend to superficial venous thrombosis (SVT) and thrombophlebitis if an infection has been superimposed.1–4
Ultrasound (US) assessment of the cannulated vein provides a useful noninvasive tool to detect the early formation of catheter-related thrombus and follow up its progression (examination of the angle between the wall of the vein and the catheter).5
El Shafei et al6 examined the use of a retrograde ventriculojugular and a retrograde ventriculosinus shunt against the direction of the blood flow. One of the advantages concluded by them was the decreased incidence of venous thrombosis that may be explained by minimized stasis around the catheter and between the catheter and the vein wall as the catheter inserted against the direction of blood flow.7
Recently, the policy of catheter replacement as clinically detected has been proved to be as safe as the routine replacement of IV catheters every 72–96 hours with many benefits.8,9 One of the factors that limits the catheter’s lifespan is catheter-related phlebitis and thrombosis that can occlude catheter and may initiate SVT.10
ICU patients need prolonged and multiple venous cannulations with consumption of accessible superficial veins. In a try to decrease the incidence of venous catheter thrombus formation and hence theoretically prolong the lifespan of indwelling venous catheters, we suggest a new technique for venous catheter insertion. There is a paucity of data regarding actual incidence of peripheral catheter-induced thrombosis.
The aim of the current study is to compare between conventional antegrade and new retrograde peripheral venous catheter insertion with regard to the incidence of catheter-related thrombus formation, onset time, and thrombus propagation time needed by the recently formed thrombus to reach catheter tip by daily real-time US examination.
The hypothesis of the study is peripheral venous catheters inserted in an opposite direction to blood flow (retrograde cannulation) can reduce the incidence of thrombus formation between the catheter and the wall of a vein, prolong the onset time of thrombus formation, delay thrombus propagation by extending the time needed by the recently formed thrombus to reach the catheter tip, and hence theoretically increase catheters’ lifespan (indwell time) when compared with antegrade insertion.
This monocentric, nonblinded, prospective observational trial was initiated in April 2015 and ended on December 2015 at Beni-Suef University Hospital in the surgical ICU that contains a total of 24 beds (8 beds for intensive and 16 beds for intermediate care). The study obtained approval by the research and ethics committee of the faculty of medicine at Beni-Suef University (FMBSU-ERC). The study has been registered at the Pan African Clinical trial registry (PACTR201504001083838). Informed consent was obtained from patients or those who are responsible for them if they were unable to give consent. We assessed 45 patients for eligibility; 5 patients were excluded (3 patients developed impairment of renal functions and 2 patients required therapeutic anticoagulation). The study included 40 patients. Inclusion criteria included the following: American Society of Anesthesiologists physical status I–IV, age 18–60 years, cephalic veins anterior to posterior dimension of not less than (2.0 mm) measured by US in both upper arms at the middle of the expected length part of vein to be occupied by the catheter using the machine’s inbuilt calipers, expected ICU stay of more than 10 days, and both forearms accessible for cannulation. Exclusion criteria included are as follows: age <18 or >60 years, thrombophilia (Factor C deficiency, Factor S deficiency, Factor 5 Leiden mutation, antithrombin III deficiency, antiphospholipid syndrome, and prothrombin 20210A mutation), thrombocytopenia with platelets counts <100,000/mL, hypocoagulable state with prothrombin concentration <40% and international normalized ratio >2, inaccessible upper arms for cannulation and presence of any contraindication to vascular cannulation in the upper limb such as skin infection, arteriovenous shunt, previous mastectomy or diagnosed SVT, and in case a hematoma or transfixion of cannulated vein occurred during insertion. Also, patients who developed impairment of renal function (elevated serum creatinine level >1.5 the baseline or urine output <0.5 mL/kg for 6 hours)11 or required higher doses of anticoagulation (eg, therapeutic doses for deep vein thrombosis) were excluded from the study.
Every recruited patient received 2 peripheral venous cannulas (BD Venflon™, Becton, Dickinson and Company, Oxford, UK) of the same caliber and length (18 G, 45 mm) in the upper limbs. The cannulation site in both arms was similar (cephalic vein), in the distal one-third of the forearm away from the wrist joint to avoid any angulation. One cannula was inserted in the direction of blood flow (antegrade cannula) with the insertion point 2 fingers above the left wrist. The other cannula was inserted in the opposite direction of blood flow (retrograde cannula) with the tip of the catheter to lay one finger above the right wrist (Figure 1). A complete aseptic condition was ensured by the use of sterile draping, sterile gloves and sterile dressing, and skin decontamination by alcohol solution with 5% chlorhexidine before skin puncture. Both venous catheters for the same patient were performed by the same operator.
The day of catheter insertion was recorded as day 0 for the lifespan of the vascular catheter and the hour of insertion was also recorded for every patient. Infusion of a Ringer acetate solution at a fixed rate of 20 mL/h was used in each IV catheter. Additional fluid administration or any medications were administered through a third IV catheter away from the study catheters. The third IV line was either an antecubital catheter in the nondominant upper limb or central venous catheter in the right internal jugular vein according to the patient’s needs. All patients included in the study received a prophylactic anticoagulation in the form of 40 mg/day subcutaneous enoxaparin (clexane, Clexane® Syringes, SANOFI, Guildford, Surrey, UK). Catheter site dressings were inspected daily at the same hour and changed immediately if damp, loosened, or soiled with subsequent assessment of both antegrade and retrograde cannulas visually for any redness, swelling, asking the patients about pain and tenderness if possible, and by US imaging with an eZono 3000 portable US system (eZono AG, Spitzweidenweg, Germany) using a L3-12 Linear Array Transducer (3-12 MHz). Only one experienced vascular sonographer performed the US examination daily to unify assessor.
The criteria of thrombus formation were defined as the presence of noncompressible echo dense material viewed at the angle between the catheter and the wall of the vein with the lack of flow (nonpulsatile and nonphasic flow). After thrombus formation at the angle between the catheter and the vein wall, its propagation to the catheter tip is followed up by daily ultrasonic assessment. Also, clinical signs of SVT and catheter occlusion as the appearance of redness, swelling, pain, and tenderness at the site of cannula insertion were recorded. Once the thrombus approached the catheter’s tip or appearance of clinical signs of catheter failure, the catheter was removed with subsequent compression for at least 15 minutes. The day at which the catheter removed was used as the lifespan of the catheter. Any complications such as thrombophlebitis, SVT, deep vein thrombosis, and pulmonary embolism were recorded and treated according to the local protocol. The primary outcome of the current study is the onset of thrombus formation viewed at the angle between the catheter and the wall of the vein as detected by US (time until the first appearance of a new incompressible echo dense shadow at the angle between the catheter and the wall of the vein). The secondary outcomes include the time needed by the recently detected thrombus to reach the catheter tip with or without catheter failure, clinical signs of SVT (redness, swelling, pain, and tenderness), and catheter occlusion (catheter failure).
Statistical Methods and Analysis
The 2 techniques of antegrade and retrograde cannulation were compared via paired t tests as each patient served as his or her own control. Normality of data distribution (within each technique and within demographic characteristics of the cohort) was examined using the Kolmogorov–Smirnov test. The normally distributed continuous variables (age, weight, height, cephalic vein dimensions, and APACHE II [Acute Physiology and Chronic Health Evaluation II] scores) were expressed as mean ± standard deviation. The nonnormally distributed continuous variables (the onset time of thrombus formation and the time needed by the newly formed thrombus to reach catheter tip) were expressed as median (interquartile range [IQR, range]) and compared for significance using Wilcoxon signed-ranks test. Categorical variables (sex and causes of ICU admission) were expressed as numbers and percentages. A P value <.05 was considered statistically significant. Statistical Package for the Social Sciences (SPSS Inc, Chicago, IL) version 16 software was used for statistical analyses of these data.
Sample size estimation depended on the data obtained from the first 9 patients of this cohort, which revealed a mean difference between the matched pairs of the onset time of thrombus formation of antegrade and retrograde cannulas (primary outcome) to be 2.49 days, standard deviation of the difference to be 4.21, and the clinical effect size d was calculated to be 0.59. Assuming α error = 0.05 (2-tailed) and β error = 0.1, a sample size of 33 patients will have the power of 90% to detect difference between the paired measurements of onset time of thrombus formation (primary outcome) of antegrade and retrograde cannula. We assume that a 30% difference in the onset time of thrombus formation (primary outcome) between both techniques of cannulas insertion could be clinically and statistically of significant importance. Enrollment of 45 patients was done to account for possible dropouts. Sample size estimation was done by G*Power software version 3.1.7 (Institute of Experimental Psychology, Heinrich Heine University, Dusseldorf, Germany).
Insertion of cannulas in the intended site position was achieved successfully in all patients. Noncompressible echodense material (thrombus) could be detected in all patients in relation to either antegrade or retrograde catheters (incidence of thrombus formation is 100% in both techniques). All catheters were functioning until removed (no reported catheters failure) depending on the thrombus propagation time to reach the catheter tip. Demographic characteristics of patients, their different causes of ICU admission, and incidence of thrombus formation are listed in Table 1. The onset time of thrombus formation between the catheter and the wall of a vein was significantly longer in the retrograde catheters with median time (IQR [range]) 6 days (5–6.75 [4–8]) with 95% confidence interval (CI) of 5.58–6.42 than in the antegrade catheters with median time (IQR [range]) 3 days (3–4 [2–5]) with 95% CI of 2.76–3.24 with P value <.001.
Thrombus propagation time, which is the time needed by the recently detected thrombus to reach the catheter tip determined by US with or without catheter failure, was significantly longer in the retrograde catheters with median time (IQR [range]) 9 days (8–9 [7–10]) with 95% CI of 8.76–9.24 when compared with the antegrade catheters with median time (IQR [range]) 4 days (4–5 [3–6]) with 95% CI of 3.76–4.24 with P value <.001. There were no reported clinical signs of thrombophlebitis, SVT, deep vein thrombosis, or pulmonary embolism in either retrograde or antegrade catheters until discharge from the ICU.
We aimed primarily in our study to evaluate the effect of retrograde peripheral venous cannulation on incidence and onset of thrombus formation and secondarily to assess thrombus propagation time, which is the time needed by the recently formed thrombus to reach the catheter tip by using the retrograde cannulation technique for peripheral veins when compared with routine antegrade cannulation. Our results revealed that the method of retrograde venous catheter cannulation did not decrease the incidence of thrombus formation but resulted in a double increase in the time onset of thrombus formation between the vein wall and the catheter and also double increase in the time of thrombus propagation, which is the time needed by the recently formed thrombus to reach the catheter tip by using real-time US examination. There are no data provided on the specific criteria determining the lifespan endpoint, ie, proportion of cannulas failing to function versus clot reaching the cannula tip and the significance of such clot in relation to cannula function, infection, and embolism remains unclear.
The need for imaging in the diagnosis of peripheral catheter-related thrombosis is crucial because most of the cases are asymptomatic. Bedside color duplex sonography was selected for detection of SVT because it is noninvasive and easy to perform with high sensitivity (93%) and specificity (93%).3
Our research depended on the physical principle presented by McCabe and Smith7 and used by El Shafei et al6 to explain the delay in clot formation around the retrograde catheter by the creation of an impact pressure zone at the tip of the retrograde catheter and at the angle between the catheter and the wall of the vein. These impact pressure zones are contrary to the wake effect zones with a static flow that is created in the same areas (tip of the catheter and the angle between the catheter and the wall of a vein) in the case of antegrade catheter insertion. By creating impact zones and avoiding wake effect zones,7 blood stasis and thrombosis are decreased and delayed. This hypothesis was also documented clinically6 and experimentally.7
The 2 main problems associated with IV catheters, which limit catheters’ lifespans, are catheter-related bloodstream infection and catheter-related phlebitis. IV catheter-related bloodstream infection is a serious complication. Fortunately, its incidence is low (0.1% of IV catheters or 0.5 per 1000 catheter days).12 The vein wall irritation (phlebitis) is the most common cause of IV catheter failures and replacements as a result of stasis, trauma, and thrombosis; its primary manifestations are pain, swelling, redness, occlusion, and a palpable venous cord.10
Both problems (infection and thrombophlebitis) were the bases of the debate between the 2 policies for catheter replacement (routine replacement or as clinically indicated replacement). The U.S. Centers for Disease Control and Prevention (CDC) guidelines released in 2011 entitle clinically indicated replacement of IV catheters in adults as an unresolved issue, signifying that more research is required.2
Rickard and colleagues8 reported that the effectiveness of routine replacement of IV catheter after 72 to 96 hours is not well established and added that this policy increases health care costs, staff workload, and requires patients to experience repeated invasive procedures. They reported a rate of 7% phlebitis in both IV catheter replacement policies and concluded that replacement of catheters as clinically indicated is of equal benefit to routine catheter replacement with negligible risk of catheter-related bloodstream infection (0.03%), local infection, and catheter tip colonization. In line with these findings, Webster and colleagues9 in their systematic review and meta-analysis concluded that although routine replacement of IV catheters is thought to reduce the risk of phlebitis and bloodstream infection, there is no evidence to support changing catheters every 72 to 96 hours and they added that health care organizations might consider shifting to a strategy whereby catheters are replaced only if clinically indicated.
The longer dwell time increases daily phlebitis risk in a linear rather than exponential manner (the more the total days of an IV catheter, the higher the overall risk of phlebitis, but the later days of the catheters’ lifespan do not carry a greater risk of phlebitis than the earlier days).13–15
The findings of the well-designed sufficiently powered 5 controlled randomized studies done between 2007 and 2012 with sample size ranging from 200 to 3283 subjects have supported clinically indicated removal of IV catheters as a safe alternative to routine replacement.8,16–19
The current real-world practice is to replace IV catheter on a clinical basis (21%–62%) rather than routine replacement after 72 to 96 hours according to CDC guidelines, and this is the result of actual practical judgments (ie, poor veins, treatment soon to be completed, and patients’ discomfort) rather than policy violation.8,20–25
Our current study examined a new technique aimed to decrease catheter-related thrombosis, which is a sequel of both phlebitis and stasis.
The primary outcome of our study is to assess the effect of the catheter insertion direction on thrombogenicity, which is particularly important in certain patients such as those with long hospital stay, difficult cannulation, unavailable limb for cannulation (eg, breast surgery or arteriovenous fistula), patients with hypercoagulable states (eg, cancer patients, postsurgical or parturients) and patients with relative or absolute contraindications to central venous cannulation (eg, coagulopathy, infection at insertion site). These patients can get benefit from prolonging catheter-related thrombosis onset and also delaying thrombus propagation by prolonging the time needed by the recently formed thrombus to reach the catheter tip. Theoretically, the indwelling time of the retrograde catheters could be extended markedly because thrombosis is an important factor that limits the catheter’s lifespan.
Reducing peripheral venous catheter-related thrombosis has 2 main benefits: first, theoretically increasing the catheter lifespan that is beneficial in many patients; and second, reducing the risk of SVT and thromboembolism.26
Upper limb deep vein thrombosis is an increasingly critical clinical issue that may lead to pulmonary embolism. Pulmonary embolism can occur in up to 30% of patients with upper limb deep vein thrombosis.27 Pain, swelling, the superior vena cava syndrome, and loss of vascular access can complicate deep venous thrombosis.28
Although deep venous thrombosis in the upper limb usually complicates peripherally inserted central venous catheters, it may also be a result of superficial thrombosis caused by peripheral venous catheters, hence the importance of monitoring and preventing SVT.30–34
On the one hand, Periard and colleagues34 reported an incidence of asymptomatic SVT with a peripheral catheter to be 37.9%, detected by the US after catheter removal; this high incidence could be attributed to a small catheter-to-vein ratio. On the other hand, Leung et al3 determined peripheral catheter-related SVT incidence to be 9.25%; they reported that it was asymptomatic and also diagnosed as an US finding after catheter removal.
US imaging not only assesses the onset and progression of thrombus formation but also gave some observations that may provide an idea about a possible effect of the direction of the blood flow (which may be related to the direction of cannulation) on thrombus formation (Figures 2–4). Retrograde catheters resulted in a thrombus characterized by a relatively concave-free surface (Figure 2) before the extension to the catheter’s tip. Thrombi around retrograde catheters were also marked by a relatively sharp free surface when reaching the tip of the catheter (Figure 3). This sharp free surface may indicate that the direction of blood flow limits the thrombus formation and extension. Antegrade catheters resulted in more extensive thrombi beyond the tip of the catheter with almost a convex free surface of the thrombus (Figure 4).
The difference in the morphology of the free surface of the thrombus in retrograde cannulation can be explained by the effect of opposing the flow of blood that delays and limits thrombus formation and propagation, respectively.
We assumed that a thrombus reaching the tip of catheter is an endpoint to remove the catheter because, theoretically, the extension of the thrombus beyond the tip can lead to a free-floating thrombus and possibility of embolism or may give rise to extension of thrombosis on the wall of vein and SVT34 with or without occlusion of the catheter. The current study is the first to use US assessment of catheter-related thrombosis as a measure to remove the catheter, because previously replacement of the catheter was either done routinely (every 72–96 hours) or as clinically indicated.8 The fear of increased risk of infection by increasing the indwelling time is controversial because it has been denied by both Lai et al35 and Rickard et al,8 who also denoted decreased cost as a result of prolonging indwell time without increased risk of phlebitis.
Although our study did not report any complications as phlebitis related to either antegrade or retrograde catheters, these results do not exclude the potential for complications and notifies the need for further studies with a larger sample size to detect any possible complications. Absent complications also may be explained by bias and the increased care (meticulous antisepsis) given to patients included in the study because blinding was absent.
The absence of blinding in our study was partly compensated by the use of a tool for monitoring and comparison between retrograde and antegrade catheters, which are real-time US imaging. The potential for investigator bias to influence the results was partly avoided by predetermining definitive criteria for assessment of thrombosis including the onset of thrombus formation and considering the extension of the thrombus to the tip of the catheter as an endpoint to remove the catheter.
We were keen to unify the factors that may affect our results as the cannulated vein, the site of cannulation, the diameter and the type of venous catheter, the type and rate of the fluid infused, the type and dose of anticoagulation, the absence of coagulopathy, inserting the venous catheters by the same operator, and US assessment by the same sonographer. This helped to ensure that we examined the effect of the retrograde technique on the onset of the thrombus formation and propagation by assessing the time needed by the recently formed thrombus to reach the catheter tip with minimal interference from other potential factors.
It is the first time to test this simple technique in practice that is used daily in almost all patients. The retrograde vascular access technique represents a simple modification that can offer a better outcome with regard to time until the onset of thrombus formation, the rate of thrombus propagation, and subsequently may increase the indwelling time of the catheter and reducing the cost.
Because the time interval of US assessment is 24 hours, in some cases of retrograde catheters, the thrombus extended beyond the tip. We did not assess the cannulated vein by US after catheter removal, but clinical follow-up until discharge from the ICU did not reveal any signs of SVT or thrombophlebitis.
We still consider our results observational and preliminary because the “objective” method of evaluation does not necessarily validate the results or objectively evaluate all variables. More studies on the novel technique are required because a cohort of 40 patients may not be enough to warrant absolute efficacy and/or safety.
Limitations and Future Plans
The routine replacement of peripheral venous catheter every 48 to 96 hours is still a mandatory policy in some medical centers according to the CDC guidelines despite that the policy is not adequately justified based on the risk of catheter-related infection. We could not blind the interpreter of the US through taking images by 1 sonographer and interpreting images by another sonographer as the diagnosis of the thrombus required testing of compressibility and assessment of flow, which needs real-time assessment by an experienced vascular sonographer. Venography,36 which is superior to sonography, was not accessible as transportation carried a risk for some of the studied patients. The anteroposterior dimensions of vessels were not measured at the junction site across the cannula groups, which can have another explanation for the results by variation in diameter owing to variation in entry site of cannulas.
In future studies, we firmly recommend performing proximal clot assessment, especially in retrograde catheters, because it was not conducted in the current trial. Different solutions, infusion rates, and pressures should be examined for the effect of their use with the new technique of cannulation and also US assessment of the cannulated vein after removal of the catheter is better to be performed for full addressing of clot formation in future studies.
Retrograde cannulation did not decrease the incidence of thrombus formation, but significantly increased the onset time until thrombus formation and prolonged the time needed by the newly formed thrombus to reach the catheter tip when compared with antegrade cannulation.
- Thrombosis complicates peripheral venous catheter in critical care patients.
- Inserting a catheter against the direction of blood flow has been used efficiently in ventriculojugular shunts with decreased stasis and thrombosis between the catheter and the wall of the vessel.
- We compared between antegrade and retrograde insertion of peripheral venous catheters in ICU patients.
- We concluded that retrograde cannulation did not decrease the incidence of thrombus formation, but significantly increased the onset time until thrombus formation and prolonged the time needed by the newly formed thrombus to reach the catheter tip when compared with antegrade cannulation.
Name: Ahmed Abdelaal Ahmed Mahmoud, MD.
Contribution: This author is the designer of the study, and helped with clinical cases, primary draft writing, final draft writing and has approved the final manuscript.
Name: Hassan Ismail El-Shafei, MD.
Contribution: This author is the inventor of the background idea (US Patent No. 7998103).
Name: Hany Mahmoud Yassin, MD.
Contribution: This author helped with data analysis and final draft writing and has approved the final manuscript.
Name: Mohamed Adly Elramely, MD.
Contribution: This author helped with ultrasound imaging and has approved the final manuscript.
Name: Mohamed Mohamed Abdelhaq, MD.
Contribution: This author helped with clinical cases and has approved the final manuscript.
Name: Hany Wafiq El Kady, MD.
Contribution: This author helped with clinical cases and data collection and has approved the final manuscript.
Name: Wael Nabil Fahemy Awada, MD.
Contribution: This author helped with clinical cases and data collection and has approved the final manuscript.
This manuscript was handled by: Richard C. Prielipp, MD.
1. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2009;49:1–45.
2. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis. 2011;52:e162–e193. doi:10.1093/cid/cir257.
3. Leung A, Heal C, Banks J, Abraham B, Capati G, Pretorius C. The incidence of peripheral catheter-related thrombosis in surgical patients. Thrombosis. 2016;2016:6043427.
4. Katz SC, Pachter HL, Cushman JG, et al. Superficial septic thrombophlebitis. J Trauma. 2005;59:750–753.
5. Srikar Adhikari. The Point of care Ultrasound diagnosis of peripheral vein septic thrombophlebitis in the emergency department. J Emerg Med. 2013;44:183–184.
6. El-Shafei IL, El-Shafei HI. The retrograde ventriculovenous shunts: the El-Shafei retrograde ventriculojugular and ventriculosinus shunts. Pediatr Neurosurg. 2010;46:160–171.
7. McCabe WL, Smith JC. Fluid flow phenomenon. Unit Operations of Chemical Engineering. 2nd ed. In: New York, NY: McGraw Hill; 47–1967:67.
8. Rickard CM, Webster J, Wallis MC, et al. Routine versus clinically indicated replacement of peripheral intravenous catheters: a randomised controlled equivalence trial. Lancet. 2012;380:1066–1074.
9. Webster J, Osborne S, Rickard CM, New K. Clinically-indicated replacement versus routine replacement of peripheral venous catheters. Cochrane Database Syst Rev. 2015;8:CD007798. doi:10.1002/14651858.CD007798.1
10. Di Nisio M, Peinemann F, Porreca E, Rutjes AW. Treatment for superficial infusion thrombophlebitis of the upper extremity. Cochrane Database Syst Rev. 2015;11:CD011015. doi: 10.1002/14651858.CD011015.pub2.
11. José AL, Sofia J. The RIFLE and AKIN classifications for acute kidney injury: a critical and comprehensive review. Clin Kidney J. 2013;6:8–14.
12. Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream infection in adults with different intravascular devices: a systematic review of 200 published prospective studies. Mayo Clin Proc. 2006;81:1159–1171.
13. Grüne F, Schrappe M, Basten J, et al. Phlebitis rate and time kinetics of short peripheral intravenous catheters. Infection. 2004;32:30–32
14. Cornely OA, Bethe U, Pauls R, Waldschmidt D. Peripheral Teflon catheters: factors determining incidence of phlebitis and duration of cannulation. Infect Control Hosp Epidemiol. 2002;23:249–253.
15. Bregenzer T, Conen D, Sakmann P, Widmer AF. Is routine replacement of peripheral intravenous catheters necessary? Arch Intern Med. 1998;158:151–156.
16. Webster J, Lloyd S, Hopkins T, Osborne S, Yaxley M. Developing a research base for Intravenous Peripheral cannula re-sites (DRIP trial). A randomised controlled trial of hospital in-patients. Int J Nurs Stud. 2007;44:664–671.
17. Webster J, Clarke S, Paterson D, et al. Routine care of peripheral intravenous catheters versus clinically indicated replacement: randomised controlled trial. BMJ. 2008;337:a339. Available at: http://www.ncbi.nlm.nih.gov/pubmed/18614482
. Accessed May 7, 2016.
18. Van Donk P, Rickard CM, McGrail MR, Doolan G. Routine replacement versus clinical monitoring of peripheral intravenous catheters in a regional hospital in the home program: a randomized controlled trial. Infect Control Hosp Epidemiol. 2009;30:915–917.
19. Rickard CM, McCann D, Munnings J, McGrail MR. Routine resite of peripheral intravenous devices every 3 days did not reduce complications compared with clinically indicated resite: a randomised controlled trial. BMC Med. 2010;8:53.
20. Zhang L, Gowardman J, Rickard CM. Impact of microbial attachment on intravascular catheter-related infections. Int J Antimicrob Agents. 2011;38:9–15.
21. Schultz AA, Gallant P. Evidence-based quality improvement project for determining appropriate discontinuation of peri pheral intravenous cannulas. Evid Based Nurs. 2005;8:8. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15688482
. Accessed May 7, 2016.
22. Palese A, Cassone A, Kulla A, et al. Factors influencing nurses’ decision-making process on leaving in the peripheral intravascular catheter after 96 hours: a longitudinal study. J Infus Nurs. 34:319–326.
23. Bravery K, Dougherty L, Gabriel J, Kayley J, Malster M, Scales K. Audit of peripheral venous cannulae by members of an i.v. therapy forum. Br J Nurs. 2006;15:1244–1249.
24. Johansson ME, Pilhammar E, Willman A. Nurses’ clinical reasoning concerning management of peripheral venous cannulae. J Clin Nurs. 2009;18:3366–3375.
25. Webster J, Osborne S, Rickard C, Hall J. Clinically-indicated replacement versus routine replacement of peripheral venous catheters. Cochrane Database Syst Rev. 2010;3:CD007798. doi:10.1002/14651858.CD007798.pub2.
26. Kooij JD, van der Zant FM, van Beek EJ, Reekers JA. Pulmonary embolism in deep venous thrombosis of the upper extremity: more often in catheter-related thrombosis. Neth J Med. 1997;50:238–242.
27. Van Canneyt K, Kips J, Mareels G, Baert E, Van Roost D, Verdonk P. Experimental & numerical modeling of the ventriculo-sinus shunt (EL Shafei shunt). J Bio Eng. 2008; 222:455–464.
28. Hylton VJ, Samuel ZG. Upper-extremity deep vein thrombosis. Circulation. 2002;106:1874–1880.
29. Gaitini D, Beck-Razi N, Haim N, Brenner B. Prevalence of upper extremity deep venous thrombosis diagnosed by color Doppler duplex sonography in cancer patients with central venous catheters. J Ultrasound Med. 2006;25:1297–1303.
30. Litzendorf ME, Satiani B. Superficial venous thrombosis: disease progression and evolving treatment approaches. Vasc Health Risk Manag. 2011;7:569–575.
31. Jeanneret C, Brunner S. Superficial venous thrombosis. A review. Hautarzt. 2012;63:609–615.
32. Wichers IM, Di Nisio M, Büller HR, Middeldorp S. Treatment of superficial vein thrombosis to prevent deep vein thrombosis and pulmonary embolism: a systematic review. Haematologica. 2005;90:672–677.
33. Bauersachs RM. Diagnosis and treatment of superficial vein thrombosis. Hamostaseologie. 2013;33:232–240.
34. Periard D, Monney P, Waeber G, et al. Randomized controlled trial of peripherally inserted central catheters vs. peripheral catheters for middle duration in-hospital intravenous therapy. J Thromb Haemost. 2008;6:1281–1288.
35. Lai KK. Safety of prolonging peripheral cannula and i.v. tubing use from 72 hours to 96 hours. Am J Infect Control. 1998;26:66–70.
36. Harigai M, Okada T, Umeoka S, et al. Non-contrast-enhanced MR venography of the upper limb: a comparative study of acquisitions with fresh blood imaging vs. time-of-flight methods. Clin Imaging. 2012;36:496–501.