Peripheral Vein Thrombophlebitis in the Upper Extremity: A Systematic Review of a Frequent and Important Problem

Journal Pre-proof

Peripheral Vein Thrombophlebitis in the Upper Extremity – a Systematic Review of a Frequent and Important Problem Shu Yun Heng M.D. , Robert Tze-Jin Yap F.A.M.S. , Joyce Tie M.B.B.S. , Duncan Angus McGrouther F.R.C.S. PII: DOI: Reference:

S0002-9343(19)30809-5 https://doi.org/10.1016/j.amjmed.2019.08.054 AJM 15405

To appear in:

The American Journal of Medicine

Please cite this article as: Shu Yun Heng M.D. , Robert Tze-Jin Yap F.A.M.S. , Joyce Tie M.B.B.S. , Duncan Angus McGrouther F.R.C.S. , Peripheral Vein Thrombophlebitis in the Upper Extremity – a Systematic Review of a Frequent and Important Problem, The American Journal of Medicine (2019), doi: https://doi.org/10.1016/j.amjmed.2019.08.054

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.

Clinical Significance  Prevention remains the cornerstone of minimising the risk of phlebitis  A holistic approach with focus on catheter insertion and care, patient comorbidities, healthcare factors should be employed  Closer monitoring should be instituted for vulnerable patients with intercurrent illness, immunocompromised states, comorbidities such as diabetes mellitus, malignancy, previous thrombophlebitis, burns  Catheter replacement should be based on clinical indication rather than routine replacement for prevention  Splintage should be considered in adult practice

1

Peripheral Vein Thrombophlebitis in the Upper Extremity – a Systematic Review of a Frequent and Important Problem Shu Yun Heng*, M.D., Robert Tze-Jin Yap*, F.A.M.S., Joyce Tie*, M.B.B.S., Duncan Angus McGrouther*, F.R.C.S. *Department of Hand and Reconstructive Microsurgery, Singapore General Hospital, 20 College Rd, Singapore 169856 Corresponding Author Prof. Duncan Angus McGrouther Address: 20 College Rd, Singapore 169856 Email: [email protected] Telephone: +6563214588 Word count: 2993 Keywords: thrombophlebitis, risk factors, patient-related, upper extremity, peripheral vein Running head: Peripheral vein thrombophlebitis in the upper extremity Declaration of Interests: None. Funding: None. All authors had access to the data and a role in writing the manuscript.

2

Abstract Background The acceptable incidence of thrombophlebitis following intravenous cannulation is 5% as recommended by the Intravenous Nurses Society guidelines, but publications have reported startling figures between 20% to 80%. Given the frequency of intravenous lines, this presents a potential clinical problem. We aimed to determine the predisposing patient, catheter and healthcare-related factors of peripheral vein thrombophlebitis in the upper extremity. Methods In this systematic review, we used a comprehensive search strategy to identify risk factors of thrombophlebitis from inception to May 20, 2019. Studies reporting risk factors of peripheral vein thrombophlebitis of adult patients admitted to hospital and receiving an intravenous cannulation were included. Quality of Prognostic Studies (QUIPS) tool was used in the assessment for risk of bias to determine the study quality. Results Of the 6910 studies initially identified, 25 were eligible for inclusion. Qualitative syntheses revealed that patient-related factors that confer a higher risk included intercurrent illness, immunocompromised state, comorbidities such as diabetes mellitus, malignancy, previous thrombophlebitis, burns, and higher haemoglobin levels. Catheter-related risk factors included catheter size, duration, and site of insertion. Intravenous antibiotics and potassium chloride predisposed to thrombophlebitis. Cannulation by an intravenous therapy team and more nursing care were associated with a decreased risk. A p-value<0.5 was considered to be statistically significant. Conclusion Recognition of the predisposing factors would allow for targeted strategies to aid in the prevention of this iatrogenic infection, which may include closer monitoring of patients who are identified to be vulnerable. Based on this systematic review, we developed an algorithm to guide clinical management. Further research is warranted to validate this algorithm. Funding None.

3

Introduction

Thrombophlebitis is a preventable iatrogenic infection that may stem from intravenous cannulation. Intravenous cannulation is a routine clinical procedure frequently performed in hospitals in order to deliver medications and fluids for therapeutic purposes. The consequences include increased healthcare costs, increased duration of hospitalisation by 7 to 14 days and also increased morbidity and mortality. 1 The additional cost of therapy ranged from $3000 to $56167 for each episode of catheter-related bloodstream infection.1 Life-threatening complications may ensue, for example bacteremia leading to septic pulmonary emboli and acute endocarditis.2 Thrombophlebitis may present with typical clinical manifestations of infection at the site of cannula insertion, or may masquerade as a benign condition in septic patients, thereby precluding prompt intervention.3 In one Cochrane review, the rate of peripheral vein phlebitis ranged between 20% to 80%, which far exceeded the acceptable incidence of 5% or less as recommended in the Intravenous Nurses Society guidelines.4 Although thrombophlebitis is a frequently encountered, potentially preventable clinical problem, it has received limited study. A review of the literature reveals that qualitative analyses of patient-related risk factors have been rare.5 Current and previous research work are mainly focused on preventive strategies and predisposing catheter-related factors.5 Similarly, preventive strategies recommended by the Centers for Disease Control and Prevention (CDC) have been primarily focused on the care of the intravenous catheter. 6 These include the duration of the replacement of catheters, frequency of inspection of the site of catheter insertion, securement with a transparent dressing, and the indication for the removal of the catheter with the development of phlebitis. Elucidation of patient-related risk factors has the potential to devise targeted strategies to guide prevention and clinical management. This may also improve current clinical practice guidelines. There is currently no consensus on the optimal management of peripheral vein thrombophlebitis of the upper extremity according to a 2015 Cochrane review.7

We performed a systematic review to determine the risk factors for peripheral vein thrombophlebitis in adult inpatients receiving a peripheral intravenous catheter in the upper extremity. Thrombophlebitis was identified clinically and defined as the presence of two or more of the following signs or symptoms on examination of the catheter insertion site or adjacent vein which include: pain, tenderness, erythema, swelling, purulence, and a palpable venous cord.8-10 This is in accordance with the widely used Visual Infusion Phlebitis scale in the diagnosis of thrombophlebitis.11 This review examines patient-related risk factors in detail, and evaluates catheter-related and healthcare-related risk factors. We have developed an algorithm guiding clinical management based on the findings of the systematic review as well as clinical practice guidelines focused on prevention and management strategies.

Methods

Search strategy and selection criteria

4

We systematically searched academic databases of Pubmed, Medline, and Embase for randomised controlled trials, cohort studies, prospective cohort studies, case control studies, cross-sectional studies, review articles, case reports, conference proceedings, and theses published between the inception of each database and 20 May 2019. The study conforms to the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) guidelines. Some of the search terms included: risk factors phlebitis, infusion phlebitis risk, peripheral intravenous catheter infection risk, and septic phlebitis (exact search terms included in Appendix I). The references of the retrieved articles were screened for relevant studies. Web-based platforms including Cochrane library, Google searches, United States Centers for Disease Control and Prevention guidelines, Infectious Diseases Society of America were also searched for grey literature. Data that were specific to peripheral vein thrombophlebitis and catheter-related infections were included. This study only included studies that reported on risk factors of peripheral vein thrombophlebitis in the upper extremity in adult patients. We excluded studies that were non-English, not conducted in the adult population, not done on humans, and studies that did not have a full-text. The adult population was defined as those who were aged 19 years and above in Pubmed and Medline, and were aged 18 years old and above in Embase. Studies that focused on central venous catheter, peripherally inserted central catheters, septic pelvic thrombophlebitis, Lemierre’s syndrome, and venous thrombosis were also excluded. In assessing the risk of bias of individual studies, the Quality in Prognostic Studies tool was employed.12 The flowchart of study selection can be found in Figure 1. The study protocol can be found in appendix II. Role of the funding source There was no funding source for this study. Results The search of published studies produced 6910 articles. 42 duplicates were removed, leaving 6868 potential relevant full-text articles as shown in the study selection in Figure 1. Amongst the 6868 potential full-text articles, 6754 were excluded as 2550 did not focus on peripheral vein thrombophlebitis, 1317 were not written in English, 2683 were not performed in the adult population, and 204 were studies conducted in animals. After reviewing the full-text articles, 25 journal articles reporting risk factors for thrombophlebitis were included; There were 4 articles that were randomised controlled trials, and 21 articles that were observational studies which were cohort studies and a descriptive comparative study. 90 articles that were not relevant, with narrow scope, conference proceedings, thesis, case reports, case series studies, cross sectional studies and review studies were excluded. The risk of bias across studies is low. A summary of the study-level, patient-level characteristics and assessment of risk of bias of each data source is provided in Table 1.8, 13-35 A summary of the risk factors for peripheral vein thrombophlebitis in the upper extremity can be found in Table 2 and Table 3. 8, 1335

Patient-related risk factors

5

The patient-related factors that confer a higher risk of peripheral vein thrombophlebitis include female gender8,13,15,31-32,35, immunocompromised states19, intercurrent illness8,13,17,21,30, comorbidities17 such as diabetes mellitus18,21, malignancy19, previous thrombophlebitis8, burns21, and higher haemoglobin levels25. The presence of intercurrent illness is a factor contributing to the development of thrombophlebitis. 8,13,17,21,30 Intercurrent illness may refer to a concurrent pneumonia, abdominal, genitourinary, skin and also an intravenous catheter-related infection.8,13,17,21,30 The higher risk could be due to chemical irritation of the endothelium from the use of intravenous antibiotics.23,33 The seeding of the catheter tip from the septic emboli from the focus of the infection may predispose a patient to thrombophlebitis. 36 Intravenous catheter-related infection may increase the risk, which can be explained by the colonisation of the catheter with skin flora that confers a six-fold increase in the risk of peripheral vein thrombophlebitis. 8 Individuals with diabetes mellitus are at an increased risk. 18,21 While Saji et al did not establish any association between diabetes mellitus and thrombophlebitis, the study found that the incidence of developing thrombophlebitis is 40% more in diabetics than in non-diabetics.37 In Monreal et al’s study, diabetes was not a risk factor.25 Diabetes mellitus is associated with endothelial dysfunction and is an inflammatory state.38 A higher incidence in diabetic patients may be attributable to the endothelial damage induced by diabetes mellitus.37

Individuals with haematological, lymphoreticular, disseminated malignancies of all systems and solid tumors are associated with a greater risk.19 Previous thrombophlebitis and burns increases this risk.8 Immunocompromised hosts are susceptible to thrombophlebitis. 39-40 These individuals are unable to mount a response to infections, and infection is one of the known mechanism for phlebitis. 41 Further research could be performed to investigate the role of end-stage renal failure, which has yet to be explored in current studies. Patients with end-stage renal failure are in an immunocompromised state due to the upregulation of the nuclear factor-kb, and this can make them susceptible to infection and thrombophlebitis.42 There exists an underlying host susceptibility towards peripheral vein thrombophlebitis, yet this concept has remained ambiguous thus far.19 In a systematic review by Tagalakis et al, the potential role of hypercoagulability, venous thrombosis and thrombophilia to explain underlying host susceptibility was proposed.4 It was raised in the same review that peripheral vein thrombophlebitis involves both inflammation and thrombus formation which resulted from catheterisation of the vein. 4 Pertaining to thrombus formation, a recent study suggested ultrasonographic evidence for thrombosis as a possible causal factor, and a recent metaanalysis demonstrated the benefits of heparin as therapeutic for thrombophlebitis.43-44 Hence, it was suggested that inherited thrombophilia, deep and superficial venous thrombosis, hypercoagulable states may explain the biologic vulnerability to thrombophlebitis. Patients with higher haemoglobin (Hb) are at a greater predisposition to thrombophlebitis.25 Patients with Hb levels between 12.5g/dl to 13.9g/dl have a greater risk compared with Hb of less than 10.5g/dl.25 A similar relationship is noted for patients with Hb more than 13.9g/dl compared with Hb levels of less than 10.5g/dl.25

6

The positive association between higher Hb and venous thromboembolism in the general population may explain the greater risk.45 Furthermore, in the Normal Hematocrit Trial, patients with end-stage renal failure on haemodialysis with Hb of approximately 14g/dl had a higher risk of thrombosis comparing with Hb of 10g/dl.46 Given that higher Hb is associated with a greater risk of thrombophlebitis and also thrombosis, this seems to support the hypothesis of thrombosis or hypercoagulability as a possible causal factor of thrombophlebitis particularly in patients with end-stage renal failure.

Catheter-related and healthcare-related risk factors Catheter-related risk factors for peripheral vein thrombophlebitis include catheter size 13,17,24,29,32, duration8,1718,20,28-29

, site8,23,28,31-35, location of insertion18,21-22, material8,16, type of infusate15,20-23,30-31,33-34, quality of vein24

and the number of catheters inserted30. Cannulation by an intravenous therapy team13-14, more nursing care27 and admission in specialised hospitals27 are significant factors associated with a decreased risk. A larger catheter is associated with thrombophlebitis. 13,17,24,29,32 This could be due to mechanical trauma to the vessel intima caused to the veins by insertion of the larger gauge catheter, which results in inflammatory changes such as intimal hyperplasia, luminal narrowing and poor circulation.47-48 The CDC recommends a change in catheter site every 72 to 96 hours.6 This is based on three studies between 1983 to 1998 which demonstrated that catheters placed in-situ for greater than 48-96 hours increased the risk of thrombophlebitis.6 This is supported by findings in our study.8,17-18,20,28-29 However, there has been differing literature.45-50 In a multi-centered, non-blinded randomised equivalence trial, the incidence of phlebitis was 7% between patients whose peripheral venous catheters were replaced when there were clinical manifestations of complications such as phlebitis, infiltration, occlusion, accidental removal, or suspected infection related to the catheter and those whose catheters were replaced routinely every 72-96 hours. 47 In two randomised controlled trials48-49, a Cochrane review50 and a systematic review45, they similarly did not demonstrate any significant differences between clinically-indicated and routine replacement of peripheral venous catheters. Furthermore, frequent changing of catheter sites can increase healthcare costs.50 This seems to suggest that catheters should be replaced according to clinical indication rather than routine replacement as a preventive strategy.

In terms of site of cannulation, the forearm or antecubital fossa seems to be at a greater risk of thrombophlebitis as compared to the wrist or dorsum of hand.8,30,33-34 The antecubital fossa may be prone to thrombophlebitis as it is a mobile joint that is susceptible to motion.28 This motion is postulated to traumatise the vessel intima, causing inflammation and subsequently thrombophlebitis.18,20,40 Nonetheless, there have been studies that demonstrated that the wrist or dorsum of hand has up to a three-fold increase in risk to peripheral vein thrombophlebitis as compared to the forearm or antecubital fossa.22,28,32 A similar explanation may be offered for the wrist as it is a mobile joint.18,20,40 The choice of cannulation at a site with bruised veins predisposes to thrombophlebitis.24 There appears to be a greater risk in catheters inserted in the emergency room as compared to the wards.18,21-22 This may be attributable to defects in aseptic technique during insertion of the catheter in the emergency setting

7

which led to introduction of the skin flora intravenously. 6 The compliance to aseptic technique should be emphasised in view that Staphylococcus aureus, which is a common skin flora, is the predominant causative organism.6 This may explain why individuals with a greater number of catheters inserted are prone to thrombophlebitis.30 Similarly, the expertise of healthcare workers has been implicated whereby cannulation with an intravenous therapy team is associated with a lower risk.13 Less experienced individuals may require multiple attempts before succeeding, and this can potentially lead to thrombophlebitis with the introduction intravenously of bacteria colonising the skin. The type of infusate determine the susceptibility to thrombophlebitis. 15,20-23,30-31,33-34 Intravenous antibiotics21,23,33 such as augmentin31, aminoglycoside31, dicloxacillin33-34, cloxacillin33-34, erythromycin33, benzylpenicillin33, and cefuroxime33 have been associated with an increased risk potentially due to the presence of microparticulates in the infusate5. Intravenous medications15,22,35 such as potassium chloride23 and hypertonic solutions20 can predispose to thrombophlebitis. The administration of corticosteroids decreases this risk. 33 While chemotherapy has not been identified in this review, other studies have proposed this as a risk factor.7-8 Chemotherapy can cause severe venous inflammation, and thus predisposes to thrombophlebitis.8 The material of catheter may influence the development of thrombophlebitis. 8,16 Tetrafluoroethylenehexafluoropropylene (FEP-Teflon) catheters may cause thrombophlebitis more often as compared with Polyurethane (PEU) vialon.8 Tetrafluoroethylene-hexafluoropropylene (Teflon) catheters have a greater association with thrombophlebitis as compared with steel needles. 16 Assessment of thrombophlebitis A high index of suspicion for treatment should be maintained. This is because patients with suppurative thrombophlebitis can have a very benign clinical presentation, and may not often present with clinical manifestations of infection at the catheter site.54 The assessment of thrombophlebitis is also complicated by the fact that current measures of phlebitis are highly observer dependent. 55 To allow for reliable and prompt identification of thrombophlebitis clinically, current assessment tools could be revised to include an objective pictorial correlation illustrating the different stages of thrombophlebitis.

Prevention and Management Strategies The occurrence of thrombophlebitis is affected by patient-related, catheter-related and healthcare-related factors.56 There is a need for education of healthcare personnel with regards to aseptic technique, compliance and competence.6 These are the areas to focus on to prevent thrombophlebitis. Although there is convincing evidence that intercurrent pathologies8,13,15,17-19,21,25,30-32,35 play an important role in the development of thrombophlebitis, patient-related factors have yet to be included in current clinical practice guidelines.2,6 Studies on catheter-related bloodstream infections report that closer surveillance of susceptible individuals and compliance to aseptic technique decrease the incidence of thrombophlebitis.57-58

8

By recognising the predisposing factors, closer monitoring and more meticulous medical care of high risk individuals can be instituted. Vulnerable patients include immunocompromised individuals19 or those with comorbidities17 such as malignancy19, diabetes mellitus18,21 and intercurrent illness8,13,17,21,30. When these individuals develop pain and erythema or swelling at catheter site, medical therapy with antimicrobial treatment should be considered, and an early surgical referral made if refractory to therapy. In terms of surgical management, Baker et al recommends that the excision of the involved vein54 should be complemented with the ligation of the surrounding veins and drainage of the corresponding pus 59-60. Secondary closure of the wound should be performed when appropriate.2,54

Discussion

Prevention remains the cornerstone of minimising the occurrence of this healthcare-related infection. To the best of our knowledge, this systematic review is the first to comprehensively review patient-related risk factors for peripheral vein thrombophlebitis. Based on the best available evidence, we have developed an algorithm employing a holistic approach to prevention and management with integration of risk factors related to catheter insertion and care, patient comorbidities, and other healthcare factors. This can be found in Figure 2.2,6-7,54 In our algorithm, we propose that closer monitoring and greater clinical suspicion of the need for treatment57-58 should be instituted for vulnerable patients. Important factors include intercurrent illness8,13,17,21,30, immunocompromised states19, comorbidities17 such as diabetes mellitus18,21, malignancy19, previous thrombophlebitis8, burns21 and higher haemoglobin levels25. The practice of replacement of catheters should be based on clinical indications rather than routine replacement every 72-96 hours as a preventive strategy.48-53 When catheter removal is based on clinical indication, they remained intact longer and had fewer complications. Moreover, there was decreased equipment requirement and staff time was reduced by 40%, with a projected annual saving in one series of $203380.80.61 Above all this preserves what may be scarce veins in patients with long term illness and avoids the pain of catheter reinsertion. As such, the conventional thinking of replacing peripheral venous catheters routinely for prevention of thrombophlebitis may need to be revisited. However, this change is predicated on the basis that the clinical manifestation of phlebitis can be recognised by healthcare personnel in a standardised manner. There may be a need to revise the current clinical assessment tools for thrombophlebitis to include more objective measurements in view of the fact that current tools have high observer-dependent variability.55 Given that the site of insertion appears to influence the development of thrombophlebitis particularly at the joints8,18,20,22,28,30,32-34,40, cannulation at the wrist or antecubital fossa should be avoided and splinted if cannulated. Splintage is a common practice in Children’s Hospitals and should be considered in adult practice. Healthcare-related factors are pivotal in prevention, as collaborative efforts by healthcare workers including clinicians, nurses and administrators are required.62 Cannulation by an intravenous therapy team13-14, more

9

nursing care27 and specialised hospitals27 are associated with a decreased risk. Lapses in infection control remain prevalent and present a potential problem on patient safety.63 Publications have reported concerning figures from 16 to 81% for the levels of compliance to hand hygiene among hospitals.63 Given that staphylococcus aureus, which is a common skin flora, is the predominant causative organism, compliance to aseptic technique should be emphasized during the catheter insertion.6 This study included both randomised controlled trials and non-randomised observational studies to obtain a complete picture of the risk factors for thrombophlebitis. Inclusion of non-randomised observational studies is necessary because there is a multitude of risk factors which have been studied in literature, but there were only few randomised controlled trials. In the selection of risk factors in the observational studies, this review only included those that were obtained from multivariate logistic regression analysis that adjusted for confounding factors. Due to the lack of studies reporting risk factors for peripheral vein thrombophlebitis in the upper extremity, there were only few patient-related risk factors elucidated. Further research is warranted to investigate other potential comorbidities such as chronic kidney disease and thrombophilia as risk factors. A registry recording all intercurrent pathologies of patients with thrombophlebitis in the upper extremity could be created. Our proposed algorithm will require further validation. More studies could be performed to evaluate cannulation at the joints as a risk factor and the role of splinting. This will allow targeted interventions to complement current clinical practice guidelines in the prevention of this iatrogenic infection. Contributors HSY, RTJY, JT and DAM conceived the idea of the study developed the protocol. HSY and DAM conducted the literature search, HSY, RTJY, JT and DAM selected the studies, extracted the studies meeting the inclusion criteria, and synthesised the data. HSY wrote the first draft of the paper. RTJY, JT and DAM critically revised successive drafts of the paper and approved the final version. DAM supervised the overall work and is the guarantor of the review. All authors reviewed the study findings and read and approved the final version before submission. Declaration of interests We declare no competing interests. References 1 Raad I, Hanna H, Maki D. Intravascular catheter-related infections: advances in diagnosis, prevention, and management. Lancet Infect Dis 2007; 7: 645–657. 2 Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: update by the Infectious Diseases Society of America. Clin Infect Dis 2009; 49: 1–45.
 3 Crowley AL, Peterson GE, Benjamin DK Jr, et al. Venous thrombosis in patients with short- and long-term central venous catheter-associated Staphylococcus aureus bacteremia. Crit Care Med 2008; 36: 385–90. 4 Zheng G, Yang L, Chen H, Chu J, Mei L. For prevention and treatment of infusion phlebitis. Cochrane Database Syst Rev 2014; 6: CD009162. 5 Tagalakis V, Kahn SR, Libman M, Blostein M. The epidemiology of peripheral vein infusion

10

thrombophlebitis: a critical review. Am J Med 2002; 113: 146–151. 6 O'Grady NP, Alexander M, Burns LA, et al. Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis 2011; 52: 23–26. 7 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. 8 Maki DG, Ringer M. Risk factors for infusion-related phlebitis with small peripheral venous catheters. Ann Intern Med 1991; 114: 845–854. 9 Monreal M, Benjamin O, Rodriguez N, et al. Infusion phlebitis in post-operative patients: when and why. Haemostasis 1999; 29: 247–254. 
 10 Hershey CO, Tomford JW, McLaren CE, et al. The natural history of intravenous catheter associated phlebitis. Arch Intern Med. 1984; 144: 1373–1375. 
 11 Jackson A. Infection control – a battle in vein: infusion phlebitis. Nurs Times 1998; 94: 67, 71. 12 Hayden JA, van der Windt DA, Cartwright JL, Côté P, Bombardier C. Assessing bias in studies of prognostic factors. Ann Intern Med 2013; 158: 280–286. 13 Wallis MC, McGrail M, Webster J, et al. Risk factors for peripheral intravenous catheter failure: a multivariate analysis of data from a randomized controlled trial. Infect Control Hosp Epidemiol 2014; 35: 63–68. 14 Soifer NE, Borzak S, Edlin BR, Weinstein RA. Prevention of peripheral venous catheter complications with an intravenous therapy team: a randomized controlled trial. Arch Intern Med 1998; 158: 473–477. 15 Sherertz R, Stephens J, Marosok R, et al. The Risk of Peripheral Vein Phlebitis Associated with Chlorhexidine-Coated Catheters A Randomised, Double-Blind Trial. Infect Control and Hosp Epidemiol 1997; 18: 230–236. 16 Tully JL, Friedland GH, Baldini LM, Goldmann DA. Complications of intravenous therapy with steel needles and Teflon catheters. A comparative study. Am J Med 1981; 70: 702–706. 17 Simin D, Milutinović D, Turkulov V, Brkić S. Incidence, severity and risk factors of peripheral intravenous cannula-induced complications: An observational prospective study. J Clin Nurs 2019; 28: 1585–1599. 18 Furtado C. Incidence and predisposing factors of phlebitis in a surgery department. Br J Nurs 2011; 20: 16– 25. 19 Tager IB, Ginsberg MB, Ellis SE, et al. An epidemiologic study of the risks associated with peripheral intravenous catheters. Am J Epidemiol 1983; 118: 839–851. 20 Uslusoy E, Mete S. Predisposing factors to phlebitis in patients with peripheral intravenous catheters: a descriptive study. J Am Acad Nurse Pract 2008; 20: 172–80. 21 Nassaji-Zavareh, Ghorbani R. Peripheral intravenous catheter-related phlebitis and related risk factors. Singapore Med J 2007; 48: 733–736. 22 Lee WL, Chen HL, Tsai TY, et al. Risk factors for peripheral intravenous catheter infection in hospitalized patients: a prospective study of 3165 patients. Am J Infect Control 2009; 37: 683–686. 23 Salgueiro-Oliveira A, Veiga P, Parreira P. Incidence of phlebitis in patients with peripheral intravenous catheters: The influence of some risk factors. Aust J Adv Nurs 2012; 30: 32–39. 24 Marsh N, Webster J, Larson, Cooke M, Mihala G, Rickard CM. Observational Study of Peripheral Intravenous Catheter Outcomes in Adult Hospitalized Patients: A Multivariable Analysis of Peripheral Intravenous Catheter Failure. J Hosp Med 2018; 13: 83–89.

11

25 Monreal M, Quilez F, Rey-Joly C, et al. Infusion phlebitis in patients with acute pneumonia: a prospective study. Chest 1999; 115: 1576–80. 26 Arias-Fernández L, Suérez-Mier B, Martínez-Ortega MC, Lana A. Incidence and risk factors of phlebitis associated to peripheral intravenous catheters. Enferm Clin 2015; 27: 79–86. 27 Palese A, Ambrosi E, Fabris F, et al. Nursing care as a predictor of phlebitis related to insertion of a peripheral venous cannula in emergency departments: findings from a prospective study. J Hosp Infect 2016; 92: 280–286. 28 Cicolini G, Manzoli L, Simonetti V, et al. Phlebitis risk varies by peripheral venous catheter site and increases after 96 hours: a large multi-centre prospective study. J Adv Nurs 2014; 70: 2539–2549. 29 Lipsky BA, Peugeot RL, Boyko EJ, Kent DL. A Prospective Study of Staphylococcus Aureus Nasal Colonization and Intravenous Therapy-Related Phlebitis. Arch Intern Med 1992; 152: 2109–2112. 30 Braga LM, Parreira PM, Oliveira SS, et al. Phlebitis and infiltration: vascular trauma associated with the peripheral venous catheter. Revista Latino-Americana de Enfermagem 2018; 26: e3002. 31 Mestre RG, Berbel BC, Tortajada LP, et al. Assessing the influence of risk factors on rates and dynamics of peripheral vein phlebitis: an observational cohort study. Med Clin (Barc) 2012; 139: 185–191. 32 Cicolini G, Bonghi AP, Di Labio L, Di Mascio R. Position of peripheral venous cannulae and the incidence of thrombophlebitis: an observational study. J Adv Nurs 2009; 65: 1268–1273. 33 Lanbeck P, Odenholt I, Paulsen O. Antibiotics differ in their tendency to cause infusion phlebitis: a prospective observational study. Scand J Infect Dis 2002; 34: 512–519. 34 Lanbeck P, Odenholt I, Paulsen O. Dicloxacillin: A Higher Risk than Cloxacillin for Infusion Phlebitis. Scand J Infect Dis 2003; 35: 397–400. 35 Abolfotouh MA, Salam M, Bani-Mustafa AB, White D, Balkhy HH. Prospective study of incidence and predictors of peripheral intravenous catheter-induced complications. Ther Clin Risk Manag 2014; 10: 993–1001. 36 Duma RJ, Warner JF, Dalton HP. Septicemia from intravenous infusions. N Engl J Med 1971; 284: 257–260. 37 Saji J, Korula SV, Mathew A, Mohan L. The incidence of thrombophlebitis following the use of peripheral intravenous cannula in post-operative patients A prospective observational study. J Dent Med Sci 2015; 14: 1–4. 38 Tabit CE, Chung WB, Hamburg NM, Vita JA. Endothelial dysfunction in diabetes mellitus: molecular mechanisms and clinical implications. Rev Endocr Metab Disord 2010; 11: 61–74. 39 Goldmann DA, Pier GB. Pathogenesis of infections related to intravascular catheterization. Clin Microbiol Rev 1993; 6: 176–92. 40 Nyika ML, Mukona D, Zvinavashe M. Factors contributing to phlebitis among adult patients admitted in the medical-surgical units of a central hospital in Harare, Zimbabwe. J Infus Nurs 2018; 41: 96–102. 41 Malach T, Jerassy Z, Rudensky B, et al. Prospective surveillance of phlebitis associated with peripheral intravenous catheters. Am J Infect Control 2006; 34: 308–312. 42 Kato S, Chmielewski M, Honda H, et al. Aspects of immune dysfunction in end-stage renal disease. Clin J Am Soc Nephrol 2008; 3: 1526–1533. 43 Everitt NJ,Krupowicz DW, Evans JA, Mc Mahon MJ. Ultrasonographic investigation of the pathogenesis of infusion thrombophlebitis. Br J Surg 1977; 84: 642–645. 
 
 44 Randolph AG, Cook DJ, Gonzales CA, Andrew M. Benefit of heparin in peripheral venous and arterial catheters: systematic review and meta-analysis of randomised controlled trials. BMJ 1998; 316: 969–975. 


12

45 Braekkan SK1, Mathiesen EB, Njølstad I, Wilsgaard T, Hansen JB. Hematocrit and risk of venous thromboembolism in a general population. The Tromso study. Haematologica 2010; 95: 270–275. 46 Besarab A, Bolton WK, Browne JK, et al. The effects of normal as compared with low haematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 1998; 339: 584–590. 47 Stokowski G, Steele D, Wilson D. The use of ultrasound to improve practice and reduce complication rates in peripherally inserted central catheter insertions: final report of investigation. J Infus Nurs 2009; 32: 145–155. 48 Chang WP, Peng YX. Occurrence of Phlebitis: A Systematic Review and Meta-analysis. Nurs Res 2018; 67: 252–260. 49 Brezenger T, Conen D, Sakman P, Widmer AF. Is routine replacement of peripheral intravenous catheters necessary? Arch Intern Med 1998; 158: 151–156. 50 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. 51 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. 52 Rickard CM, McCann D, Munnings J, et al. 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. 53 Webster J, Osborne S, Rickard CM, New K. Clinically-indicated replacement versus routine replacement of peripheral venous catheters. Cochrane Database Syst Rev 2015; 14: CD007798. 54 Baker CC, Petersen SR, Sheldon GF. Septic phlebitis: a neglected disease. Am J Surg 1979; 138: 97–103. 55 Ray-Barruel G, Polit DF, Murfield JE, Rickard CM. Infusion phlebitis assessment measures: a systematic review. J Eval Clin Pract 2014; 20: 191–202. 56 Zingg W, Pittet D. Peripheral venous catheters: an under-evaluated problem. Int J Antimicrob Agents 2009; 34: S38–42. 57 Mehta Y, Gupta A, Todi S, et al. Guidelines for prevention of hospital acquired infections. Indian J Crit Care Med 2014; 18: 149–163. 58 Gahlot R, Nigam C, Kumar V, et al. Catheter-related bloodstream infections. Int J Crit Illn Inj Sci 2014; 4: 162–167. 59 Crane C. Venous interruption for septic thrombophlebitis. N Engl J Med 1960; 262: 947–951. 60 Vie-DuPont J, Cormier JM, LeCompte Y. Ligation of veins in suppurative thrombophlebitis secondary to venous catheterization. Surg Gynecol Obstet 1974; 138: 662. 61 Maier D. To Replace or Not to Replace? Replacing Short Peripheral Catheters Based on Clinical Indication. J Infus Nurs 2014; 42: 143–148. 62 Braithwaite J. Changing how we think about healthcare improvement. BMJ 2018; 361: k2014. 63 Burke JP. Infection control – a problem for patient safety. N Engl J Med 2003; 348: 651–656.

13

Figures legends

Figure 1. Flowchart of study selection.

14

Figure 2. An Algorithm for prevention and management strategies for peripheral vein thrombophlebitis in the upper extremity. *Taken from reference 4 United States Centers for Disease Control and Prevention guidelines for the prevention of intravascular catheter-related infections, 2011. **Taken from reference 5 Cochrane review on treatment of superficial infusion thrombophlebitis of the upper extremity. ***Taken from reference 1 clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 Update by the Infectious Diseases Society of America.

15

Population Patients older than 18 years old without granulocytopenia who received a peripheral intravenous catheter Adult medical and surgical patients with a peripheral intravenous venous catheter with greater than 4 days of expected use All inpatients receiving medical service

Study type Randomised, controlled trial

Dates 1991

Randomised, controlled trial

Randomised, controlled trial

Patients older than 18 years old with expected duration of catheterisation of more than 24 hours Adult patients admitted to the medical, surgical and gynaecologic ward at Beth Israel Hospital with a peripheral venous cannula Patients older than 18 years old indicated for intravenous therapy through a peripheral cannula expected to last longer than 24 hours Patients older than 18 years old admitted to the general surgery department, and had a peripheral cannula receiving infusions or intravenous drugs for at least 24 hours

al

Maki et al (1991)

Wallis (2014)

et

al

Soifer (1998)

et

al

Sherertz (1997)

et

al

Tully et al (1981)

Simin (2019)

Furtado (2011)

Tager (1983)

et

et

et

al

al

Participants 1054

Bias Low

May 2018 to September 2009

3283

Low

441

Low

Randomised, double-blind trial

3 month period before 1998 6 month period in 1990

226

Low

Randomised prospective cohort

September 1978 to April 1979

954

Low

Prospective cohort

April 2015 to September 2015

368

Low

Prospective cohort

15 October to 30 November 2010

171

Low

Patients admitted to 10 non-Federal hospitals in the State of Rhode Island

Prospective cohort

March 1980 to February 1982

3094

Low

Uslusoy (2007)

et

al

Patients admitted to the general surgery unit who had intravenous catheterization and were receiving a drug treatment or infusion

Descriptive, comparative

10 January to 30 April 2006

355

Low

Nassaji (2007)

et

al

Patients admitted to medical and surgical wards in Semnan, Iran Adult patients more than or equal to 20

Prospective cohort

April 2003 to February 2004 1 January 2005 to 20

300

Low

3165

Low

Lee et al (2009)

Prospective cohort

16

SalgueiroOliveira et (2012) Marsh (2018)

et

al

al

Monreal (1999)

et

al

Monreal (1999)

et

al

Arias-Fernández et al (2015)

Palese (2016)

et

al

Cicolini (2014)

et

al

Lipsky (1992)

et

al

Braga (2018)

et

al

years old admitted to medical or surgical wards Patients with peripheral intravenous catheter hospitalised for 6 weeks Patient over 18 years old with a peripheral intravenous catheter inserted 24 hours, and who were able to provide written informed consent Patients with acute pneumonia receiving intravenous therapy at an internal medicine department of a tertiary teaching hospital Patients admitted to the Department of General Surgery who were scheduled to have a peripheral intravenous catheter inserted for routine antibiotic therapy Adult inpatients that had at least one peripheral venous catheter in an intentional sample from the floors of the Hospital Universitario Central of Asturias Urgent cases who had a peripheral venous catheter inserted in the emergency department and were subsequently admitted to a medical unit for at least 24 hours, who were willing to participate in the study, were included Adult patients admitted to the inpatient ward Adult patients admitted to the inpatient wards at Seattle (Wash) Veterans Affairs Medical Centre Adult patients more than 18 years old admitted to the medical clinic of a

August 2005 Prospective cohort

30 January 2010 to 12 March 2010

1244

Low

Prospective cohort

October 2014 to December 2015

1000

Low

Prospective cohort

1999

766

Low

Prospective cohort

Over a period of 3 years before 1999

400

Low

Prospective cohort

1 February 2015 to 15 February 2015

105

Low

Prospective cohort

7 month period between 2012 to 2013

1262

Low

Prospective cohort

January 2012 to June 2012 10 week period in 1991

1498

Low

350

Low

10 July 2015 to 10 September 2015

110

Low

Prospective cohort

Prospective cohort

17

hospital in the central area of Portugal Mestre Roca et al All peripheral short Prospective November 967 Low (2012) intravenous catheters cohort 2005 inserted in patients aged 18 years who were admitted to any ward G. Cicolini et al Patients admitted to Prospective 2007 427 Low (2009) the surgical and cohort internal medicine departments with a catheter from the emergency room Lanbeck et al Patients older than 17 Prospective 21 January 628 Low (2002) years old with cohort 1996 to 31 peripheral venous May 1996 catheters at the Department of Infectious Disease, Malmö University Hospital Lanbeck et al Patients older than 17 Prospective 21 January 550 Low (2003) years old with cohort 1996 to 31 peripheral venous May 1996 catheters at the Department of Infectious Disease, Malmö University Hospital Abolfotouh et at Adult patients more Prospective November 359 Low (2014) than 18 years old cohort 2012 to admitted to selected January units at King 2013 Abdulaziz Medical city with peripheral intravenous catheter insertion Table 1. A summary of the study-level and patient-level characteristics of each data source meeting inclusion criteria.

18

Patient-related factors Female gender8,13,15,31-32,35 Presence of intercurrent illness8,13,17,21,30 (including intravenous catheterrelated infection) Presence of comorbidities17 Diabetes mellitus18,21

Immunodeficiency diseases19 Malignancy19 (hematologic/ lymphoreticular, disseminated of all systems, solid tumors) Previous thrombophlebitis8 Burns21 Hb 12.5-13.9 compared to Hb<10.525 Hb>13.9 compared to Hb<10.525

Catheter-related factors Duration8,17-18,20,28-29 Longer duration>shorter duration8,17-18,20,28-29 Size13,17,24,29,32 Large bore>small bore13,17,24,29,32 Site8,23,28,31-35 Lower extremity>Lower Extremity23 Forearm/Antecubital fossa>wrist/dorsum of hand8,28,32-35 Forearm>antecubital Fossa31 Quality of insertion site24 Bruised>pristine24

Healthcare-related factors Expertise of healthcare personnel13-14 House staff>intravenous therapy team13-14 Nursing care27 Less care>more care27 Type of hospital27

nursing nursing

General>specialised 27

Location18,21-23 Emergency room>ward18,21-23 Underlying host susceptibility19 Infusate15,20-23,30-21,33-34 Antibiotics21,23,33 Hypertonic solutions20 KCl23 Continuous intravenous infusion22 Intravenous medications15,35 Augmentin, aminoglycoside, dicloxacillin, erythromycin, corticosteroids, benzylpenicillin, cefuroxime33 Dicloxacillin>cloxacillin34 Number of catheters inserted30 Greater number>smaller number30 Catheter Material8,16 Teflon catheter>steel needle16 FEP-Teflon>PEU vialon8 Table 2. Risk factors for peripheral vein thrombophlebitis in the upper extremity. Identified in prospective studies by multivariate logistic regression analysis or in randomised controlled trials. The > symbol denotes a significantly greater risk for thrombophlebitis.

19

Risk factors Patient-related factors Female gender

Presence of intercurrent illness (including intravenous catheterrelated infection)

Presence of comorbidities Diabetes mellitus Immunodeficiency diseases Malignancy (hematologic/lymphoreticular, disseminated of all systems, solid tumors) Previous thrombophlebitis Burns Hb 12.5-13.9 compared to Hb<10.5 Hb>13.9 compared to Hb<10.5 Underlying host susceptibility Catheter-related factors Duration Longer duration>shorter duration

Size Large bore>small bore

Site Lower extremities>upper extremities Forearm/Antecubital fossa>wrist/dorsum of hand

Forearm>antecubital fossa

Study

Relative Risk/Odds Ratio/Hazard Ratio (95% Confidence Interval)

Maki et al, 1991 Wallis et al, 2014 Sherertz et al, 1997 Mestre Roca et al, 2012 Cicolini et al, 2009 Abolfotouh et al, 2014 Maki et al, 1991

1.64(1.22-2.19) 1.64(1.28-2.09) 2.84(1.43-5.67) 1.46(1.01-2.14) 1.91(1.20-3.03) 2.44(1.68-3.54) 6.19(3.42-11.77)

Wallis et al, 2014

1.41(1.05-1.89)

Simin et al, 2019

1.42(1.06-1.91)

Nassaji et al, 2007

6.21(4.27-9.03)

Braga et al, 2018

1.93(1.05-3.54)

Simin et al, 2019

1.48(1.17-1.86)

Furtado et al, 2011 Nassaji et al, 2007 Tager et al, 1983 Tager et al, 1983

2.42(1.23-4.77) 7.78(4.59-13.21) 1.66(1.24-2.23) 1.66(1.24-2.23)

Maki et al, 1991 Nassaji et al, 2007 Monreal et al, 1999 Monreal et al, 1999 Tager et al, 1983

1.54(1.12-2.13) 3.96(3.26-4.82) 1.81(1.28-2.58) 2.30(1.58-3.33) 1.66(1.24-2.23)

Maki et al, 1991 Simin et al, 2019 Furtado et al, 2011 Uslusoy et al, 2007 Cicolini et al, 2014 Lipsky et al, 1992

3.97(2.34-6.76) 2.08(1.40-3.08) 2.72(1.56-4.75) 1.69(1.14-2.51) 1.05(1.03-1.07) 3.35(1.49-7.69)

Wallis et al, 2014 Simin et al, 2019 Arias-Fernández et al, 2015 Lipsky et al, 1992 Cicolini et al, 2009

1.48(1.08-2.03) 2.95(1.60-5.41) 4.35(1.25-14.29)

Salgueiro-Oliveira al, 2012 Maki et al, 1991

3.11(1.07-9.01)

et

Cicolini et al, 2014 Cicolini et al, 2009 Lanbeck et al, 2002 Lanbeck et al, 2003 Abolfotouh et al, 2014 Mestre Roca et al, 2012

3.35(1.49-7.69) 5.38(2.70-10.72)

1.67(1.14-2.44) 0.52(0.32-0.84) 0.30(0.12-0.73) 1.80(1.30-2.49) 2.60(1.11-6.07) 1.53(1.05-2.22) 1.93(1.20-3.01)

20

Quality of insertion site Bruised>pristine Location Emergency room>ward

Infusate Antibiotics

Hypertonic solutions Potassium chloride Continuous intravenous infusion Intravenous medications Augmentin Aminoglycoside

Marsh et al, 2018

1.39(1.26-3.71)

Furtado et al, 2011 Nassaji et al, 2007 Lee et al, 2009

1.88(1.04-3.41) 2.04(1.36-3.05) 1.60(1.003-2.50)

Lanbeck et al, 2002 Salgueiro-Oliveira et al, 2012 Nassaji et al, 2007 Uslusoy et al, 2007 Salgueiro-Oliveira et al, 2012 Lee et al, 2009 Sherertz et al, 1997 Abolfotouh et al, 2014 Mestre Roca et al, 2012 Mestre Roca et al, 2012 Lanbeck et al, 2002 Lanbeck et al, 2002 Lanbeck et al, 2002 Lanbeck et al, 2002 Lanbeck et al, 2002 Lanbeck et al, 2003

2.09(1.41-3.10) 1.88(1.14-3.09) 6.21(4.27-9.03) 1.83(1.16-2.89) 2.11(1.12-3.97) 3.80(1.20-11.60) 3.07(1.28-7.37) 1.55(1.07-2.24) 2.15(1.46-3.20) 2.13(1.01-4.63)

Dicloxacillin 7.17(3.72-13.82) Corticosteroids 0.48(0.34-0.63) Erythromycin 4.06(2.09-7.88) Benzylpenicillin 2.27(1.45-3.54) Cefuroxime 2.60(1.32-6.07) Dicloxacillin>cloxacillin 3.48(1.64-7.38) Number of catheters inserted Greater number>smaller number Braga et al, 2018 1.37(1.15-1.64) Catheter Material Teflon catheter>steel needle Tully et al, 1981 1.87(1.45-2.40) FEP-Teflon>PEU vialon Maki et al, 1991 0.58(0.43-0.78) Healthcare-related factors Expertise of healthcare personnel House staff>intravenous therapy Wallis et al, 2014 1.69(1.30-2.20) team Nursing care Less nursing care>more nursing care Palese et al, 2016 0.99(0.98-0.99) Type of hospital General>Specialised Palese et al, 2016 0.058(0.025-0.366) Table 3. Risk factors for peripheral vein thrombophlebitis in the upper extremity. Identified in prospective studies by multivariate logistic regression analysis or in randomised controlled trials. The > symbol denotes a significantly greater risk for thrombophlebitis. The p-value of all the above-mentioned risk factors are less than 0.5.

21