- Email: [email protected]
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters
+Model
ARTICLE IN PRESS
Enferm Clin. 2016;xxx(xx):xxx---xxx
www.elsevier.es/enfermeriaclinica
ORIGINAL ARTICLE
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters夽 Incidencia y factores de riesgo de flebitis asociadas a catéteres venosos periféricos Loreto Arias-Fernándeza,b , Belén Suérez-Miera , María del Carmen Martínez-Ortegaa , Alberto Lanab,∗ a b
Servicio de Medicina Preventiva, Hospital Universitario Central de Asturias, Oviedo, Spain Departamento de Medicina, Área de Medicina Preventiva y Salud Pública, Universidad de Oviedo, Oviedo, Spain
Received 29 December 2015; accepted 12 July 2016
KEYWORDS Phlebitis; Peripheral venous catheterization; Risk factors; Incidence; Nursing
Abstract Objective: To determine the incidence and risk factors of phlebitis associated to the care of peripheral vascular catheters (PVC). Method: Prospective cohort study at the Central University Hospital of Asturias (Spain). A total of 178 PVC were observed daily until their extraction. The incidence of phlebitis was measured using the visual infusion phlebitis scale, that distinguishes between grade I (possible phlebitis) and II (phlebitis). The independent diagnoses of phlebitis made by staff nurses were also collected. Finally, data about the insertion and the care of the PVC was also obtained. The incidence of phlebitis and the validity of the diagnoses made by staff nurses were calculated and the risk factors of phlebitis were determined by means of logistic regression. Results: 5.6% of the PVC presented phlebitis, 21.3% possible phlebitis and 11.2% had phlebitis according to nurses’ criteria. The staff nurses had a sensitivity of 100%, a specificity of 94% and a positive predictive value of 50% in the diagnosis of phlebitis. After adjusting for potential confounders, the use of an extension tube as an accessory of the PVC was an independent predictor of phlebitis (odds ratio: 4.8; P = 0.04), but a PVC size of 22/24 gauges was associated with lower phlebitis incidence (odds ratio: 0.2; P = 0.02). Conclusions: Clinical phlebitis assessment is difficult because the agreement for phlebitis diagnosis is low. To minimise the incidence of phlebitis would be recommended to choose the smallest PVC size possible and to avoid using an extension tube as an accessory of the PVC. © 2016 Elsevier Espa˜ na, S.L.U. All rights reserved.
夽 Please cite this article as: Arias-Fernández L, Suérez-Mier B, Martínez-Ortega MC, Lana A. Incidencia y factores de riesgo de flebitis asociadas a catéteres venosos periféricos. Enferm Clin. 2016. http://dx.doi.org/10.1016/j.enfcli.2016.07.008 ∗ Corresponding author. E-mail address: [email protected] (A. Lana).
http://dx.doi.org/10.1016/j.enfcle.2016.07.002 2445-1479/© 2016 Elsevier Espa˜ na, S.L.U. All rights reserved.
ENFCLE-613; No. of Pages 8
+Model
ARTICLE IN PRESS
2
L. Arias-Fernández et al.
PALABRAS CLAVE Flebitis; Cateterización venosa periférica; Factores de riesgo; Incidencia; Enfermería
Resumen Objetivo: Determinar la incidencia de flebitis y los factores de riesgo relacionados con el cuidado de los catéteres venosos periféricos (CVP). Método: Estudio de cohortes prospectivo en el Hospital Universitario Central de Asturias. Se incluyeron 178 CVP que fueron observados diariamente hasta su retirada. La ocurrencia de flebitis fue medida con la escala visual de flebitis por infusión, que permite distinguir entre grado i (posible flebitis) y grado ii (flebitis). También se recogió el diagnóstico de flebitis que realizó de forma independiente el personal de enfermería de las unidades. Por último, se obtuvo información sobre la inserción y el cuidado del CVP. Se calculó la incidencia de flebitis y la validez de los diagnósticos realizados en la unidad y, mediante regresión logística, se identificaron los factores de riesgo de flebitis. Resultados: El 5,6% de los CVP presentaron flebitis, el 21,3% posibles flebitis y el 11,2% flebitis según el criterio del personal enfermero. La sensibilidad de los profesionales para diagnosticar flebitis fue del 100%, la especificidad, del 94%, y el valor predictivo positivo, del 50%. Después de controlar los potenciales confusores, la utilización de alargadera como accesorio del CVP incrementó el riesgo de flebitis (odds ratio: 4,8; p = 0,04) e insertar un catéter de calibre 22/24 gauges lo disminuyó (odds ratio: 0,2; p = 0,02). Conclusiones: La falta de consenso sobre el diagnóstico de la flebitis dificulta su valoración por parte de los profesionales. Para minimizar la incidencia de flebitis sería aconsejable elegir el calibre de CVP más peque˜ no posible y evitar el uso de alargaderas como accesorio. © 2016 Elsevier Espa˜ na, S.L.U. Todos los derechos reservados.
What is known? Phlebitis is one of the most common complications associated with peripheral venous catheterisation. The variability in estimates makes study of a complication whose risk factors require clarification difficult.
What do we contribute? Staff nurses are very sensitive but less specific in making a diagnosis of phlebitis. The use of an extension tube as a catheter accessory and a wider bore are risk factors associated with nursing care.
Introduction The World Health Organisation estimates that 5---10% of patients suffer some form of harm while receiving care in technologically developed hospitals; this is fundamentally due to healthcare-related infections.1 Nurse-managed healthcare very often requires the insertion of a peripheral venous catheter (PVC). This is the most common invasive procedure carried out in hospitals and has an incidence varying between 70% and 80% of hospitalised patients.2,3 The procedure can compromise patient safety since it is associated with local and systemic
complications such as infections at the insertion point, phlebitis, bacteraemia and sepsis.2,3 Phlebitis symptoms are the most common complication,3,4 since, although the variability in estimates is enormous, the estimated prevalence ranges between 20% and 65% of patients with a PVC.5 Phlebitis is defined as the inflammation of a vein due to a disturbance the endothelium during or after intravenous infusion. It is characterised by the following symptoms, pain and local swelling, erythema around the puncture site or along the course of the vein, local heat, palpable venous cord accompanied or otherwise by a purulent discharge and/or fever. Venous thrombosis can occur in the most serious cases.4,6,7 The consequences of phlebitis can be major and affect both the patient and the healthcare system. Patients can experience pain and distress, anxiety, impaired venous pool, eruption of their prescribed therapy. They can require further treatment, suffer bacteraemia or require an increased hospital stay, amongst other consequences. Furthermore, phlebitis can result in increased human and material health care costs. It is estimated that PVC insertion costs around 25 dollars, not to mention the suffering caused to the patient or the time taken by the practitioner to perform the technique.8 In terms of annual morbidity, the incidence of these adverse effects is considerable due to the increasing use of PVC and the high incidence of phlebitis.2 There is abundant scientific literature regarding the risk factors and complications associated with central venous catheters. However, there is little literature on phlebitis associated with PVC, and it is less conclusive.2 According to Hadaway,9 the causes of infections and phlebitis
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters
3
No signs of phlebitis. OBSERVE the insertion site
Pain with no erythema, swelling or palpable cord in the area of function
1
Possible sign of phlebitis OBSERVE the insertion site
Pain with erythema and/or swelling with no palpable cord in the puncture site.
2
Start of phlebitis REMOVE the catheter
Pain, erythema, swelling, hardening or palpable venous cord <6cm above the insertion site
3
Mid phase of phlebitis REMOVE the catheter and assess treatment
Pain, erythema, swelling, hardening or palpable venous cord <6cm above the insertion site and/or purulence
4
Advanced stage of phlebitis. REMOVE the catheter and assess treatment
Frank venous thrombosis with all the above signs and difficult or halted infusion
5
Thrombophlebitis. REMOVE the catheter and start treatment
Moderate pain
0
Severe pain
No pain, erythema, swelling or palpable cord
No pain
Visual Infusion Phlebitis Scale
Figure 1 Visual infusion phlebitis scale. Image taken from: flebitiszero.com.17
associated with PVC still need to be clarified. Furthermore, most research has taken place in Anglo-Saxon countries such as England, Australia and the United States,10---14 where there are catheterisation specialists, therefore their outcomes and conclusions do not necessarily apply to countries where all nursing professionals are competent to insert and manage these devices, as is the case in Spain. Our aim was to measure the incidence of phlebitis and clarify the phlebitis risk factors associated with PVC insertion and care.
Method Design, population and scope of the study A prospective cohort study performed with inpatients that had at least one PVC in an intentional sample from the floors of the Hospital Universitario Central of Asturias. Inclusion criteria were, being an internal medical, medical oncology, neurosurgery or haematology department inpatient, of legal age and providing informed consent. Only PVC inserted from 1 to 15 February 2015 were included. Patients with PVC inserted in different departments and those transferred to units not included in the study were excluded in order to ensure adequate follow-up of the PVC. Given the observational nature of the study, the investigators had no role in the insertion or care of the PVC, which were inserted by nurses according to their professional criteria. The Hospital Universitario Central of Asturias has a protocol for the insertion, maintenance and removal of PVC.15 The final sample comprised 105 patients and 178 PVC (unit of analysis).
Study variables and data collection procedure The principal study variable was the grade of phlebitis. The visual infusion phlebitis scale was used, which was created by Jackson based on the Maddox scale.16 This scale was subsequently adapted and validated by Schultz and Gallant,8 and is currently one of the main consensus tools for staging phlebitis. It defines and describes 6 grades of phlebitis and offers a recommendation for action for each (Fig. 1).17 Three criteria were used in the context of this study to establish phlebitis. On the one hand, a strict criterion, where a grade II or greater on the visual infusion phlebitis scale was considered phlebitis. On the other, a wider criterion considering a grade I to be phlebitis, i.e., according to this criterion ‘‘possible phlebitis’’ was included as phlebitis. This is the most used criterion in the scientific literature.5 Finally, a removal criterion was used, i.e., phlebitis reported by the staff nurse on the floor and which had resulted in discontinuing the PVC. A series of independent variables were also collected due to their potential association with the phlebitis. The nurse in charge of the daily observation of the patient obtained part of this information directly and recorded it on a data collection sheet. It was necessary to review the electronic clinical records to gather further data. Specifically, information was obtained on (a) the patients’ clinico-epidemiological variables and drug treatment: sex, age, duration of hospital stay (<10 days, 10---20 days, >20 days), comorbidity (arterial hypertension, diabetes, obesity and/or cancer), weight loss over the past 6 months, febrile episode during admission, treatment with anticoagulants and intravenous antibiotic treatment (non-irritant or irritant), electrolytes, analgesia, blood/blood products or chemotherapy agents (non aggressive or aggressive: blistering or irritant), and
+Model
ARTICLE IN PRESS
4 (b) variables associated with the PVC and its care: indwell time (<48 h, 48---96 h, >96 h), insertion shift (morning, afternoon, or night), PVC bore (18---20G or 22---24G), insertion site (hand, forearm, flexure or arm), side (left or right), number of previous catheters during the same period of admission (none, one or two, or more), type of dressing (gauze or transparent/semitransparent), change of dressing, securing means (none, non-sterile strips or sterile strips), presence of an extension tube, use of bioconnector or infusion pump. On the first day of data collection for each PVC, only the variables relating to its insertion and securing means were recorded. On the subsequent days, each PVC was visually monitored for the onset of the signs and symptoms of phlebitis. A PVC specialist nurse, trained in using the scale, measured the phlebitis by direct, daily observation of the PVC insertion site. In parallel, the phlebitis data that the staff nurses had recorded in the patients’ electronic clinical history were gathered, as well as the remaining necessary information. Two investigators performed a separate quality control to the information achieved above.
Data analysis The incidence of phlebitis was calculated according to each criterion used. First, the incidence proportion was calculated, dividing the number of phlebitis cases by the number of PVC and then the incidence rate, dividing the number of phlebitis cases between the sum of days with PVC over the period studied and multiplying by 1000 (incidence of phlebitis per 1000 PVC days). The sample was described using absolute and relative frequencies (%) for the qualitative variables, and means and standard deviation (SD) for the quantitative variables. The validity of the clinical diagnoses of phlebitis made by nurses was studied by comparing them with the diagnoses made by the investigators using the visual infusion phlebitis scale (gold standard). To be specific, the sensitivity and specificity parameters and the predictive values were calculated using the standard formulae. Finally a multivariant analysis was carried out by binary regression logistics (probability of phlebitis according to the wider criterion) to establish the factors that significantly contributed towards its onset. Adjusted odds ratios (OR) were calculated and their 95% confidence intervals (CI95%) entering the above-mentioned independent variables in 2 steps: step 1, the clinic-epidemiological and treatment variables and step 2, the variables related to the PVC and its care. The statistical analyses were performed using the SPSS software package version 22 for Windows (IMB Corp., New York, U.S.A.). Only P values <0.05 were considered statistically significant.
Ethical considerations The Clinical Research Ethics Committee of the Principality of Asturias and the Management of the Hospital Universitario Central of Asturias approved the study.
L. Arias-Fernández et al. Table 1 Clinico-epidemiological and drug treatment features according to phlebitis (wider criterion). Phlebitis No
Yes
Female, n (%)
61 (43.6)
27 (71.1)
Age, n (%) <65 years 65---79 years ≥80 years
60 (42.9) 42 (30.0) 38 (27.1)
16 (42.1) 14 (36.8) 8 (21.1)
Duration of hospital stay, n (%) <10 days 10---19 days ≥20 days
41 (29.3) 51 (36.4) 48 (34.3)
6 (15.8) 21 (55.3) 11 (28.9)
Arterial hypertension, n (%) Diabetes, n (%) Neoplasm, n (%) Loss of weight, n (%) Fever, n (%) Anticoagulants, n (%)
53 29 52 19 5 95
14 7 11 6 2 28
Antibiotics, n (%) None Non irritant Irritant
64 (45.7) 40 (28.6) 36 (25.7)
17 (44.7) 10 (26.3) 11 (28.9)
Chemotherapy agents, n (%) None Non aggressive Aggressive
132 (94.3) 2 (1.4) 6 (4.3)
35 (92.1) 1 (2.6) 2 (5.3)
Analgesia, n (%) Blood/blood products, n (%)
67 (47.9) 13 (9.3)
18 (47.4) 3 (7.9)
(37.9) (20.8) (37.1) (13.6) (3.6) (67.9)
(36.8) (18.4) (28.9) (15.8) (5.3) (73.7)
Results Following what was defined as the strict criterion, 10 cases of phlebitis were diagnosed (5.6% of the PVC). Because each PVC had been inserted for a mean 3.4 days (SD 2.5), the incidence rate was 16.6 phlebitis per 1000 PVC days. Using the wider criteria, which considered a grade I or more to be phlebitis, the incidence proportion was 21.3% (38 cases out of 178 PVC studied) and the incidence rate of phlebitis was 63 per 1000 days with a PVC. Finally, following the removal criterion, the staff nurses diagnosed 20 cases of phlebitis (11.2% of the PVC) and the incidence rate was 33.2 phlebitis per 1000 PVC days. Considering the outcome of applying the visual infusion phlebitis scale as the gold standard, the staff nurses’ sensitivity in diagnosing phlebitis without using any scale was 100% and specificity 94%. The positive predictive value of diagnosis was 50% and the negative predictive value 100%. The clinico-epidemiological features and drug treatment of the PVC patients are shown on Table 1. The females, aged between 65 and 80 years, hospitalised for between 10
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters Table 2 Features of PVC and its care according to the phlebitis (wider criterion). Phlebitis No Indwell time, n (%) 0---2 days 3---4 days >4 days Nursing shift, n (%) Morning Afternoon Evening Bore, n (%) 18---20G 22---24G Anatomical site, n (%) Hand Forearm Flexure Arm Side, n (%) Right Left Previous catheters, n (%) None One Two or more Type of dressing, n (%) Gauze Transparent Change of dressing, n (%)
76 (54.3) 44 (31.4) 20 (14.3) 27 (19.3) 113 (80.7) 41 69 13 17
Table 3 Odds ratio (OR) and confidence intervals (CI95%) of the association between the features of the PVC and its care and the incidence of phlebitis (wider criterion). OR (IC95%)a
P
Indwell time, n (%) 0---2 days 3---4 days >4 days
1.00 3.19 (0.79---12.91) 0.50 (0.09---2.77)
0.11 0.43
Nursing shift, n (%) Morning Afternoon Evening
1.00 0.92 (0.23---3.68) 0.70 (0.11---4.36)
0.90 0.70
Bore, n (%) 18-20G 22-24G
1.00 0.23 (0.07---0.80)
0.02
Anatomic site, n (%) Hand Forearm Flexure Arm
1.00 1.42 (0.41---4.99) 2.86 (0.29---28.64) 2.34 (0.34---16.23)
0.58 0.37 0.39
Side, n (%) Right Left
1.00 0.56 (0.20---1.59)
0.28
Previous catheters, n (%) None One Two or more
1.00 0.56 (0.10---3.19) 0.44 (0.07---2.75)
0.52 0.38
Type of dressing, n (%) Gauze Transparent Change of dressing, n (%)
1.00 9.16 (0.56---150.3) 0.70 (0.21---2.29)
0.12 0.69
Securing system, n (%) None Sterile strips Non-sterile strips
1.00 0.10 (0.01---2.05) 0.07 (0.01---1.40)
0.13 0.08
Extension tube, n (%) Bioconnector, n (%) Infusion pump, n (%)
4.82 (1.05---22.13) 1.38 (0.12---16.49) 0.18 (0.02---1.63)
0.04 0.79 0.13
Yes
67 (47.9) 36 (25.7) 37 (26.4)
(29.3) (49.3) (9.3) (12.1)
71 (50.7) 69 (49.3) 23 (16.4) 38 (27.1) 79 (56.4) 39 (27.8) 101 (72.1) 58 (41.4)
12 (31.6) 15 (39.5) 11 (28.9) 27 (71.1) 8 (21.1) 3 (7.9) 15 (39.5) 23 (60.5) 10 22 2 4
(26. 3) (57.9) (5.3) (10.5)
19 (50.0) 19 (50.0) 9 (23.7) 10 (26.3) 19 (50.0) 3 (7.9) 35 (92.1) 19 (50.0)
Securing system, n (%) None Sterile strips Non-sterile strips
33 (23.6) 102 (72.9) 5 (3.6)
5 (13.2) 31 (81.6) 2 (5.3)
Extension tubing, n (%) Bioconnector, n (%) Infusion pump, n (%)
13 (9.3) 131 (93.6) 27 (19.3)
7 (18.4) 36 (94.7) 2 (5.3)
to 20 days and treated with anticoagulants, had a greater incidence of phlebitis. Similarly, at bivariant level, the features of the PVC features and its care that formed a risk profile for phlebitis were: a longer PVC indwell time---it would have been inserted in the morning---a larger bore, using a transparent dressing and any other securing means, using an extension tube as an accessory and less use of an infusion pump (Table 2). However, when a controlled multivariant analysis was performed to monitor the effect of all the variables (Table 3), the only features the PVC care associated significantly with phlebitis were: insertion of a 22 or 24 gauge PVC, which reduced the risk (OR: 0.23; CI95%: 0.07---0.80), and the PVC having an
5
a
OR adjusted by the table variables and by: sex, age (<65 years, 65---79 years or ≥80 years), duration of hospital stay (<10 days, 10---19 days or ≥20 days), arterial hypertension, diabetes, neoplasm, weight loss, fever, oral anticoagulants, intravenous antibiotics (none, non irritant or irritant), intravenous chemotherapy agents (none, non aggressive or aggressive), analgesia and blood/blood products.
extension tube, which increased the risk (OR: 4.82; CI95%: 1.05---22.13).
Discussion According to the results of our prospective study, the incidence rate of phlebitis in a Spanish reference hospital is greater than the theoretically acceptable level. Using a
+Model
ARTICLE IN PRESS
6 wider definition of phlebitis (all grade I or higher), we found that a larger PVC bore and the use of extension tubing were the only variables associated with the insertion of the PVC or with its care which significantly increased the risk of phlebitis. There is enormous disparity in the rates of phlebitis in the scientific literature, which appears to be more associated with the lack of consensus on a definition that would enable a clear diagnosis than to differences in relation to the care of PVC.18 On the one hand, the incidence of phlebitis that we found using the strict criteria was practically the same as that reported in the Iberoamerican Study on Adverse Events in Latin-American hospitals19 (5.7%) and only slightly higher than that found by Ferrete-Morales et al.2 (4.8%) prior to the implementation of a specific PVC action protocol. However, on the other hand, the incidence rate found by the COSMOS study in a third level Spanish hospital was very much higher than ours, varying between 31 and 45 cases of phlebitis per 100 PVC days.20 In any case, in most of the studies performed in Anglo-Saxon countries, phlebitis is diagnosed the minute there is pain at the insertion site (grade I), i.e., using what we call the wider criterion. In these studies, the rates of phlebitis tend to vary between 20% and 30%,21,22 our estimate also found this range. In the United States, the phlebitis rate acceptable to the Infusion Nursing Society is 5% or less in all populations,23 therefore we could consider that the incidence in Asturias’ Hospital Universitario Central is only moderately acceptable using the strict criterion, and improvable if we use the wider criterion. An examination of the incidence rate of phlebitis using the removal criterion---which constituted a measurement of the staff nurses’ clinical judgement---and comparing it with what is considered the gold standard, revealed that staff nurses have high sensitivity in diagnosing phlebitis, which results in a poor positive predictive value of their evaluations, in other words, only half of the cases where the nurses had decided to remove the PVC actually had phlebitis. In the same regard, Marsh et al.18 found a low rate of agreement using different scales for evaluating phlebitis, this result raises the question of the need for agreed valid and reliable measurements of phlebitis. Furthermore, it is well known that early removal of a PVC results in major financial losses, discomfort for patients and hinders nurses undertaking other tasks.24 With regard to nursing care that might affect the risk of phlebitis, the multivariant analysis revealed that, irrespective of the other factors, only 2 nursing decisions regarding the insertion and care of a PVC were clinically relevant: the bore and the use of tube extensions. Amongst the mechanical factors, the PVC bore is one of the most studied. In general, there is relative consensus in the scientific literature with regard to the relationship between a larger PVC bore and a greater incidence of phlebitis, possibly due to the physical trauma caused to the vein intima.25---28 For this reason, it would seem reasonable to choose the lowest possible PVC bore, although obviously as the international guidelines recommend, what it is to be used for, the risk of other complications and patient comfort should be taken into account.14 However, it is still quite routine to choose
L. Arias-Fernández et al. the largest bore that the patient can tolerate, with the futile objective of avoiding obstructions that might cause a delay in nursing tasks. Finally, the effects of extension tubes have been broadly debated, with no clear recommendations on the subject. In Spain, studies that have attempted to investigate the association between PVC extension tubes and the risk of phlebitis attribute a protective effect to their use, but we consider that these outcomes might be outdated because of the modifications that have been made in PVC materials and accessories.29 For example, the study by Rivas et al.30 established that placing extension tubing between the PVC and the 3-way key reduced the risk of phlebitis, perhaps because this reduces the traction exercised by the weight of the key on the tip of the PVC. However, 3-way keys are used so seldom nowadays that our outcomes are not comparable. In fact, none of the patients in our series had a 3-way key. More recently in a doctoral thesis, Nu˜ nezCrespo31 found no significant relationship either between the use of extension tubing and the incidence of phlebitis, but the lack of association might be due to the fact that the study lacked the statistical power necessary to make this contrast, because, in fact, in that series, only the patients with extension tubing presented phlebitis. The lack of statistical power to correctly assess the risk factors is a problem that cuts across most phlebitis studies.32 Finally, according to the results of the COSMOS study, closed systems---i.e., with extension tubing incorporated into the PVC---reduce the risk of phlebitis and costs compared to accessory extension tubing (as occurred in our study).20 Therefore, it appears reasonable to avoid the use of accessory extension tubing. In any case, we were not able to establish whether it is the extension tubing itself or accessing it while failing to respect the recommended asepsis that constitutes the risk factor. The latter is unlikely because other similar accessories would then also have been demonstrated as risk factors (e.g. bioconnectors). Our study has certain limitations. Firstly, despite having an appropriate sample size to establish the incidence of phlebitis with a margin of error of less than 3%, the sample could not be sufficient to study the effect of some potential risk factors. Secondly, the selection of the participants could also have been a source of error, as the sample was not random. Nonetheless, we tried to collect a sample from the 2 large hospital areas (a medical floor and a surgical floor) and from other more specific areas (haematology and oncology). Finally, some risk of residual confusion cannot be ruled out since there might be other important variables which have not been gathered in this study, and therefore not checked in the analysis. In conclusion, the lack of consensus on the diagnosis of phlebitis makes it difficult to assess this complication. Staff nurses tend to make a diagnosis of phlebitis and to remove the PVC before it is clinically advisable, which can reduce the real incidence of phlebitis but it can also increase patient discomfort and the costs for the system. According to the results of studying the risk factors, in order to minimise the incidence of phlebitis it would be advisable to choose the smallest bore of PVC possible and avoid the use of accessory extension tubes.
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters
Conflict of interests The authors have no conflict of interests to declare.
14.
Acknowledgements The authors would like to thank the supervisors of the hospital units included in the study for their collaboration: Eugenia García, M. Antonia Duarte, Susana Campos and Carmen Vena. We would also like to thank Julián Díaz for his participation and advice in piloting the data collection instrument. This research was conducted in the context of the Flebitis-Zero project (http://flebitiszero. com/app/).
15.
16. 17.
References 1. Alianza Mundial para la Seguridad del Paciente. La Investigación en Seguridad del Paciente [Internet]. París: Organización Mundial de la Salud; 2008. Available in: http://www.who.int/patientsafety/information centre/ documents/ps research brochure es.pdf [accessed 28.11.15]. 2. Ferrete-Morales C, Vázquez-Pérez MA, Sánchez-Berna M, Gilabert-Cerro I, Corzo-Delgado JE, Pineda-Vergara JA, et al. Incidencia de flebitis secundaria por catéter venoso de acceso periférico e impacto de un protocolo de manejo. Enferm Clin. 2010;20:3---9. 3. Pasalioglu KB, Kaya H. Catheter indwell time and phlebitis development during peripheral intravenous catheter administration. Pak J Med Sci. 2014;30:725---30. 4. Nassaji-Zavareh M, Ghorbani R. Peripheral intravenous catheter related phlebitis and related risk factors. Singap Med J. 2007;48:733---6. 5. Uslusoy E, Mete S. Predisposing factors to phlebitis in patients with peripheral intravenous catheters: a descriptive study. J Am Acad Nurs Pract. 2008;20:172---80. 6. Washington GT, Barret R. Peripheral phlebitis: a pointprevalence study. J Infus Nurs. 2012;35:252---8. 7. Kaur P, Thakur R, Kaur S, Bhalla A. Assessment of risk factors of phlebitis amongst intravenous cannulated patients. Nurs Midwifery Res J. 2011;7:106---14. 8. Gallant P, Schultz AA. Evaluation of a Visual Infusion Scale for determining appropriate discontinuation of peripheral intravenous catheters. J Infus Nurs. 2006;29: 338---45. 9. Hadaway L. Short peripheral intravenous catheters and infections. J Infus Nurs. 2012;35:230---40. 10. Lee WL, Chen HL, Tsai TY, Lai IC, Chang WC, Huang CH, 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---6. 11. Rickard CM, Webster J, Wallis MC, Marsh N, McGrail MR, French V, et al. Routine versus clinically indicated replacement of peripheral intravenous catheters: a randomised controlled equivalence trial. Lancet. 2012;380: 1066---74. 12. O’Grady NP, Alexander M, Burns A, Dellinger EP, Garland J, Heard SO, et al. Guidelines for the Prevention of Intravascular Catheter-Related Infections [Internet]. Atlanta: Centers for Disease Control and Prevention (CDC); 2011. Available in: http://www.cdc.gov/hicpac/pdf/guidelines/bsi-guidelines2011.pdf [accessed 14.10.15]. 13. Webster J, Osborne S, Rickard CM, New K. Clinically-indicated replacement versus routine replacement of peripheral
18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
28.
29.
7
venous catheters. Cochrane Database Syst Rev. 2015;8: CD007798. Loveday HP, Wilson JA, Pratt RJ, Golsorkhi M, Tingle A, Bak A, et al. epic3: National evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J Hosp Infect. 2014;86 Suppl. 1: S1---70. González E, Argüelles AR, Martínez B. Comisión de Cuidados de Enfermería. Protocolo para la inserción, mantenimiento y retirada del catéter venoso periférico [Internet]. Oviedo: Hospital Universitario Central de Asturias; 2013. Available in: http://www.hca.es/huca/web/enfermeria/html/f archivos/ Cateter%20venoso%20periferico.pdf [accessed 20.06.16]. Jackson A. Infection control-a battle in vein: infusion phlebitis. Nurs Times. 1998;94:68---71. Martínez-Ortega C, Suárez Mier B (coordinadoras). Flebitis Zero: diagnóstico de la flebitis [Internet]. Oviedo: Hospital Universitario Central de Asturias; 2015. Available in: http://flebitiszero.com/app/formacion/DiagnosticoFlebitis. html [accessed 20.06.16]. Marsh N, Mihala G, Ray-Barruel G, Webster J, Wallis MC, Rickard CM. Inter-rater agreement on PIVC-associated phlebitis signs, symptoms and scales. J Eval Clin Pract. 2015;21: 893---9. Aranaz JM, Aibar C, Larizgoitia I, Gonseth J, Colomer C, Terol E, et al. Estudio IBEAS. Prevalencia de efectos adversos en hospitales de Latinoamérica [Internet]. Madrid: Ministerio de Sanidad y Política Social; 2010. Available in: http://www.seguridaddelpaciente.es/resources/contenidos/ castellano/2009/INFORME IBEAS.pdf [accessed 03.05.15]. González López JL, Arribi Vilela A, Fernández del Palacio E, Olivares Corral J, Benedicto Martí C, Herrera Portal P. Indwell times, complications and costs of open vs closed safety peripheral intravenous catheters: a randomized study. J Hosp Infect. 2014;86:117---26. Davies J, Cotton J, Wood J, Burke D, McMahon M. Intravenous nutrition----peripheral or central administration? Br J Surg. 2003;90 Suppl. 1:86---91. Grune F, Schrappe M, Basten J, Wenchel HM, Tual E, Stutzer H. Phlebitis rate and time kinetics of short peripheral intravenous catheters. Infection. 2004;32:30---2. Infusion Nurses Society. Infusion nursing ---- standards of practice. J Infus Nurs. 2011;34 Suppl. 1:1---106. Hasselberg D, Ivarsson B, Andersson R, Tingstedt B. The handling of peripheral venous catheters----from noncompliance to evidence-based needs. J Clin Nurs. 2010;19: 3358---63. Wallis MC, McGrail M, Webster J, Marsh N, Gowardman J, Playford EG, 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---8. Saini R, Agnihotri M, Gupta A, Walia I. Epidemiology of infiltration and phlebitis. Nurs Midwifery Res J. 2011;7: 22---33. Tagalakis V, Kahn SR, Libman M, Blostein M. The epidemiology of peripheral vein infusion thrombophlebitis: a critical review. Am J Med. 2002;113:146---51. Cicolini G, Manzoli L, Simonetti V, Flacco ME, Comparcini D, Capasso L, et al. Phlebitis risk varies by peripheral venous catheter site and increases after 96 h: a large multi-centre prospective study. J Adv Nurs. 2014;70: 2539---49. López Rodríguez L, Seda Diestro J, Garrido Díaz-Malaguilla IM. Llave de tres vías con alargadera en complicaciones asociadas a la vía intravenosa. Enferm Clin. 1998;8: 104---9.
+Model
ARTICLE IN PRESS
8 30. Rivas JS, Artes JL, Arjona FJ, Carmona D, Soriano FJ, Mejías FM, et al. Flebitis traumáticas. Cómo disminuir su incidencia. Rev Enferm. 2004;27:602---6. 31. Nú˜ nez-Crespo FJ. Efectos adversos hospitalarios en dispositivos venosos periféricos: estudio de validez diagnóstica [tesis]. Madrid: Universidad Rey Juan Carlos; 2014.
L. Arias-Fernández et al. 32. Webster J, McGrail M, Marsh N, Wallis M, Ray-Barruel G, Rickard CM. Post infusion phlebitis: incidence and risk factors. Nurs Res Pract. 2015;2015:691934.
ARTICLE IN PRESS
Enferm Clin. 2016;xxx(xx):xxx---xxx
www.elsevier.es/enfermeriaclinica
ORIGINAL ARTICLE
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters夽 Incidencia y factores de riesgo de flebitis asociadas a catéteres venosos periféricos Loreto Arias-Fernándeza,b , Belén Suérez-Miera , María del Carmen Martínez-Ortegaa , Alberto Lanab,∗ a b
Servicio de Medicina Preventiva, Hospital Universitario Central de Asturias, Oviedo, Spain Departamento de Medicina, Área de Medicina Preventiva y Salud Pública, Universidad de Oviedo, Oviedo, Spain
Received 29 December 2015; accepted 12 July 2016
KEYWORDS Phlebitis; Peripheral venous catheterization; Risk factors; Incidence; Nursing
Abstract Objective: To determine the incidence and risk factors of phlebitis associated to the care of peripheral vascular catheters (PVC). Method: Prospective cohort study at the Central University Hospital of Asturias (Spain). A total of 178 PVC were observed daily until their extraction. The incidence of phlebitis was measured using the visual infusion phlebitis scale, that distinguishes between grade I (possible phlebitis) and II (phlebitis). The independent diagnoses of phlebitis made by staff nurses were also collected. Finally, data about the insertion and the care of the PVC was also obtained. The incidence of phlebitis and the validity of the diagnoses made by staff nurses were calculated and the risk factors of phlebitis were determined by means of logistic regression. Results: 5.6% of the PVC presented phlebitis, 21.3% possible phlebitis and 11.2% had phlebitis according to nurses’ criteria. The staff nurses had a sensitivity of 100%, a specificity of 94% and a positive predictive value of 50% in the diagnosis of phlebitis. After adjusting for potential confounders, the use of an extension tube as an accessory of the PVC was an independent predictor of phlebitis (odds ratio: 4.8; P = 0.04), but a PVC size of 22/24 gauges was associated with lower phlebitis incidence (odds ratio: 0.2; P = 0.02). Conclusions: Clinical phlebitis assessment is difficult because the agreement for phlebitis diagnosis is low. To minimise the incidence of phlebitis would be recommended to choose the smallest PVC size possible and to avoid using an extension tube as an accessory of the PVC. © 2016 Elsevier Espa˜ na, S.L.U. All rights reserved.
夽 Please cite this article as: Arias-Fernández L, Suérez-Mier B, Martínez-Ortega MC, Lana A. Incidencia y factores de riesgo de flebitis asociadas a catéteres venosos periféricos. Enferm Clin. 2016. http://dx.doi.org/10.1016/j.enfcli.2016.07.008 ∗ Corresponding author. E-mail address: [email protected] (A. Lana).
http://dx.doi.org/10.1016/j.enfcle.2016.07.002 2445-1479/© 2016 Elsevier Espa˜ na, S.L.U. All rights reserved.
ENFCLE-613; No. of Pages 8
+Model
ARTICLE IN PRESS
2
L. Arias-Fernández et al.
PALABRAS CLAVE Flebitis; Cateterización venosa periférica; Factores de riesgo; Incidencia; Enfermería
Resumen Objetivo: Determinar la incidencia de flebitis y los factores de riesgo relacionados con el cuidado de los catéteres venosos periféricos (CVP). Método: Estudio de cohortes prospectivo en el Hospital Universitario Central de Asturias. Se incluyeron 178 CVP que fueron observados diariamente hasta su retirada. La ocurrencia de flebitis fue medida con la escala visual de flebitis por infusión, que permite distinguir entre grado i (posible flebitis) y grado ii (flebitis). También se recogió el diagnóstico de flebitis que realizó de forma independiente el personal de enfermería de las unidades. Por último, se obtuvo información sobre la inserción y el cuidado del CVP. Se calculó la incidencia de flebitis y la validez de los diagnósticos realizados en la unidad y, mediante regresión logística, se identificaron los factores de riesgo de flebitis. Resultados: El 5,6% de los CVP presentaron flebitis, el 21,3% posibles flebitis y el 11,2% flebitis según el criterio del personal enfermero. La sensibilidad de los profesionales para diagnosticar flebitis fue del 100%, la especificidad, del 94%, y el valor predictivo positivo, del 50%. Después de controlar los potenciales confusores, la utilización de alargadera como accesorio del CVP incrementó el riesgo de flebitis (odds ratio: 4,8; p = 0,04) e insertar un catéter de calibre 22/24 gauges lo disminuyó (odds ratio: 0,2; p = 0,02). Conclusiones: La falta de consenso sobre el diagnóstico de la flebitis dificulta su valoración por parte de los profesionales. Para minimizar la incidencia de flebitis sería aconsejable elegir el calibre de CVP más peque˜ no posible y evitar el uso de alargaderas como accesorio. © 2016 Elsevier Espa˜ na, S.L.U. Todos los derechos reservados.
What is known? Phlebitis is one of the most common complications associated with peripheral venous catheterisation. The variability in estimates makes study of a complication whose risk factors require clarification difficult.
What do we contribute? Staff nurses are very sensitive but less specific in making a diagnosis of phlebitis. The use of an extension tube as a catheter accessory and a wider bore are risk factors associated with nursing care.
Introduction The World Health Organisation estimates that 5---10% of patients suffer some form of harm while receiving care in technologically developed hospitals; this is fundamentally due to healthcare-related infections.1 Nurse-managed healthcare very often requires the insertion of a peripheral venous catheter (PVC). This is the most common invasive procedure carried out in hospitals and has an incidence varying between 70% and 80% of hospitalised patients.2,3 The procedure can compromise patient safety since it is associated with local and systemic
complications such as infections at the insertion point, phlebitis, bacteraemia and sepsis.2,3 Phlebitis symptoms are the most common complication,3,4 since, although the variability in estimates is enormous, the estimated prevalence ranges between 20% and 65% of patients with a PVC.5 Phlebitis is defined as the inflammation of a vein due to a disturbance the endothelium during or after intravenous infusion. It is characterised by the following symptoms, pain and local swelling, erythema around the puncture site or along the course of the vein, local heat, palpable venous cord accompanied or otherwise by a purulent discharge and/or fever. Venous thrombosis can occur in the most serious cases.4,6,7 The consequences of phlebitis can be major and affect both the patient and the healthcare system. Patients can experience pain and distress, anxiety, impaired venous pool, eruption of their prescribed therapy. They can require further treatment, suffer bacteraemia or require an increased hospital stay, amongst other consequences. Furthermore, phlebitis can result in increased human and material health care costs. It is estimated that PVC insertion costs around 25 dollars, not to mention the suffering caused to the patient or the time taken by the practitioner to perform the technique.8 In terms of annual morbidity, the incidence of these adverse effects is considerable due to the increasing use of PVC and the high incidence of phlebitis.2 There is abundant scientific literature regarding the risk factors and complications associated with central venous catheters. However, there is little literature on phlebitis associated with PVC, and it is less conclusive.2 According to Hadaway,9 the causes of infections and phlebitis
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters
3
No signs of phlebitis. OBSERVE the insertion site
Pain with no erythema, swelling or palpable cord in the area of function
1
Possible sign of phlebitis OBSERVE the insertion site
Pain with erythema and/or swelling with no palpable cord in the puncture site.
2
Start of phlebitis REMOVE the catheter
Pain, erythema, swelling, hardening or palpable venous cord <6cm above the insertion site
3
Mid phase of phlebitis REMOVE the catheter and assess treatment
Pain, erythema, swelling, hardening or palpable venous cord <6cm above the insertion site and/or purulence
4
Advanced stage of phlebitis. REMOVE the catheter and assess treatment
Frank venous thrombosis with all the above signs and difficult or halted infusion
5
Thrombophlebitis. REMOVE the catheter and start treatment
Moderate pain
0
Severe pain
No pain, erythema, swelling or palpable cord
No pain
Visual Infusion Phlebitis Scale
Figure 1 Visual infusion phlebitis scale. Image taken from: flebitiszero.com.17
associated with PVC still need to be clarified. Furthermore, most research has taken place in Anglo-Saxon countries such as England, Australia and the United States,10---14 where there are catheterisation specialists, therefore their outcomes and conclusions do not necessarily apply to countries where all nursing professionals are competent to insert and manage these devices, as is the case in Spain. Our aim was to measure the incidence of phlebitis and clarify the phlebitis risk factors associated with PVC insertion and care.
Method Design, population and scope of the study A prospective cohort study performed with inpatients that had at least one PVC in an intentional sample from the floors of the Hospital Universitario Central of Asturias. Inclusion criteria were, being an internal medical, medical oncology, neurosurgery or haematology department inpatient, of legal age and providing informed consent. Only PVC inserted from 1 to 15 February 2015 were included. Patients with PVC inserted in different departments and those transferred to units not included in the study were excluded in order to ensure adequate follow-up of the PVC. Given the observational nature of the study, the investigators had no role in the insertion or care of the PVC, which were inserted by nurses according to their professional criteria. The Hospital Universitario Central of Asturias has a protocol for the insertion, maintenance and removal of PVC.15 The final sample comprised 105 patients and 178 PVC (unit of analysis).
Study variables and data collection procedure The principal study variable was the grade of phlebitis. The visual infusion phlebitis scale was used, which was created by Jackson based on the Maddox scale.16 This scale was subsequently adapted and validated by Schultz and Gallant,8 and is currently one of the main consensus tools for staging phlebitis. It defines and describes 6 grades of phlebitis and offers a recommendation for action for each (Fig. 1).17 Three criteria were used in the context of this study to establish phlebitis. On the one hand, a strict criterion, where a grade II or greater on the visual infusion phlebitis scale was considered phlebitis. On the other, a wider criterion considering a grade I to be phlebitis, i.e., according to this criterion ‘‘possible phlebitis’’ was included as phlebitis. This is the most used criterion in the scientific literature.5 Finally, a removal criterion was used, i.e., phlebitis reported by the staff nurse on the floor and which had resulted in discontinuing the PVC. A series of independent variables were also collected due to their potential association with the phlebitis. The nurse in charge of the daily observation of the patient obtained part of this information directly and recorded it on a data collection sheet. It was necessary to review the electronic clinical records to gather further data. Specifically, information was obtained on (a) the patients’ clinico-epidemiological variables and drug treatment: sex, age, duration of hospital stay (<10 days, 10---20 days, >20 days), comorbidity (arterial hypertension, diabetes, obesity and/or cancer), weight loss over the past 6 months, febrile episode during admission, treatment with anticoagulants and intravenous antibiotic treatment (non-irritant or irritant), electrolytes, analgesia, blood/blood products or chemotherapy agents (non aggressive or aggressive: blistering or irritant), and
+Model
ARTICLE IN PRESS
4 (b) variables associated with the PVC and its care: indwell time (<48 h, 48---96 h, >96 h), insertion shift (morning, afternoon, or night), PVC bore (18---20G or 22---24G), insertion site (hand, forearm, flexure or arm), side (left or right), number of previous catheters during the same period of admission (none, one or two, or more), type of dressing (gauze or transparent/semitransparent), change of dressing, securing means (none, non-sterile strips or sterile strips), presence of an extension tube, use of bioconnector or infusion pump. On the first day of data collection for each PVC, only the variables relating to its insertion and securing means were recorded. On the subsequent days, each PVC was visually monitored for the onset of the signs and symptoms of phlebitis. A PVC specialist nurse, trained in using the scale, measured the phlebitis by direct, daily observation of the PVC insertion site. In parallel, the phlebitis data that the staff nurses had recorded in the patients’ electronic clinical history were gathered, as well as the remaining necessary information. Two investigators performed a separate quality control to the information achieved above.
Data analysis The incidence of phlebitis was calculated according to each criterion used. First, the incidence proportion was calculated, dividing the number of phlebitis cases by the number of PVC and then the incidence rate, dividing the number of phlebitis cases between the sum of days with PVC over the period studied and multiplying by 1000 (incidence of phlebitis per 1000 PVC days). The sample was described using absolute and relative frequencies (%) for the qualitative variables, and means and standard deviation (SD) for the quantitative variables. The validity of the clinical diagnoses of phlebitis made by nurses was studied by comparing them with the diagnoses made by the investigators using the visual infusion phlebitis scale (gold standard). To be specific, the sensitivity and specificity parameters and the predictive values were calculated using the standard formulae. Finally a multivariant analysis was carried out by binary regression logistics (probability of phlebitis according to the wider criterion) to establish the factors that significantly contributed towards its onset. Adjusted odds ratios (OR) were calculated and their 95% confidence intervals (CI95%) entering the above-mentioned independent variables in 2 steps: step 1, the clinic-epidemiological and treatment variables and step 2, the variables related to the PVC and its care. The statistical analyses were performed using the SPSS software package version 22 for Windows (IMB Corp., New York, U.S.A.). Only P values <0.05 were considered statistically significant.
Ethical considerations The Clinical Research Ethics Committee of the Principality of Asturias and the Management of the Hospital Universitario Central of Asturias approved the study.
L. Arias-Fernández et al. Table 1 Clinico-epidemiological and drug treatment features according to phlebitis (wider criterion). Phlebitis No
Yes
Female, n (%)
61 (43.6)
27 (71.1)
Age, n (%) <65 years 65---79 years ≥80 years
60 (42.9) 42 (30.0) 38 (27.1)
16 (42.1) 14 (36.8) 8 (21.1)
Duration of hospital stay, n (%) <10 days 10---19 days ≥20 days
41 (29.3) 51 (36.4) 48 (34.3)
6 (15.8) 21 (55.3) 11 (28.9)
Arterial hypertension, n (%) Diabetes, n (%) Neoplasm, n (%) Loss of weight, n (%) Fever, n (%) Anticoagulants, n (%)
53 29 52 19 5 95
14 7 11 6 2 28
Antibiotics, n (%) None Non irritant Irritant
64 (45.7) 40 (28.6) 36 (25.7)
17 (44.7) 10 (26.3) 11 (28.9)
Chemotherapy agents, n (%) None Non aggressive Aggressive
132 (94.3) 2 (1.4) 6 (4.3)
35 (92.1) 1 (2.6) 2 (5.3)
Analgesia, n (%) Blood/blood products, n (%)
67 (47.9) 13 (9.3)
18 (47.4) 3 (7.9)
(37.9) (20.8) (37.1) (13.6) (3.6) (67.9)
(36.8) (18.4) (28.9) (15.8) (5.3) (73.7)
Results Following what was defined as the strict criterion, 10 cases of phlebitis were diagnosed (5.6% of the PVC). Because each PVC had been inserted for a mean 3.4 days (SD 2.5), the incidence rate was 16.6 phlebitis per 1000 PVC days. Using the wider criteria, which considered a grade I or more to be phlebitis, the incidence proportion was 21.3% (38 cases out of 178 PVC studied) and the incidence rate of phlebitis was 63 per 1000 days with a PVC. Finally, following the removal criterion, the staff nurses diagnosed 20 cases of phlebitis (11.2% of the PVC) and the incidence rate was 33.2 phlebitis per 1000 PVC days. Considering the outcome of applying the visual infusion phlebitis scale as the gold standard, the staff nurses’ sensitivity in diagnosing phlebitis without using any scale was 100% and specificity 94%. The positive predictive value of diagnosis was 50% and the negative predictive value 100%. The clinico-epidemiological features and drug treatment of the PVC patients are shown on Table 1. The females, aged between 65 and 80 years, hospitalised for between 10
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters Table 2 Features of PVC and its care according to the phlebitis (wider criterion). Phlebitis No Indwell time, n (%) 0---2 days 3---4 days >4 days Nursing shift, n (%) Morning Afternoon Evening Bore, n (%) 18---20G 22---24G Anatomical site, n (%) Hand Forearm Flexure Arm Side, n (%) Right Left Previous catheters, n (%) None One Two or more Type of dressing, n (%) Gauze Transparent Change of dressing, n (%)
76 (54.3) 44 (31.4) 20 (14.3) 27 (19.3) 113 (80.7) 41 69 13 17
Table 3 Odds ratio (OR) and confidence intervals (CI95%) of the association between the features of the PVC and its care and the incidence of phlebitis (wider criterion). OR (IC95%)a
P
Indwell time, n (%) 0---2 days 3---4 days >4 days
1.00 3.19 (0.79---12.91) 0.50 (0.09---2.77)
0.11 0.43
Nursing shift, n (%) Morning Afternoon Evening
1.00 0.92 (0.23---3.68) 0.70 (0.11---4.36)
0.90 0.70
Bore, n (%) 18-20G 22-24G
1.00 0.23 (0.07---0.80)
0.02
Anatomic site, n (%) Hand Forearm Flexure Arm
1.00 1.42 (0.41---4.99) 2.86 (0.29---28.64) 2.34 (0.34---16.23)
0.58 0.37 0.39
Side, n (%) Right Left
1.00 0.56 (0.20---1.59)
0.28
Previous catheters, n (%) None One Two or more
1.00 0.56 (0.10---3.19) 0.44 (0.07---2.75)
0.52 0.38
Type of dressing, n (%) Gauze Transparent Change of dressing, n (%)
1.00 9.16 (0.56---150.3) 0.70 (0.21---2.29)
0.12 0.69
Securing system, n (%) None Sterile strips Non-sterile strips
1.00 0.10 (0.01---2.05) 0.07 (0.01---1.40)
0.13 0.08
Extension tube, n (%) Bioconnector, n (%) Infusion pump, n (%)
4.82 (1.05---22.13) 1.38 (0.12---16.49) 0.18 (0.02---1.63)
0.04 0.79 0.13
Yes
67 (47.9) 36 (25.7) 37 (26.4)
(29.3) (49.3) (9.3) (12.1)
71 (50.7) 69 (49.3) 23 (16.4) 38 (27.1) 79 (56.4) 39 (27.8) 101 (72.1) 58 (41.4)
12 (31.6) 15 (39.5) 11 (28.9) 27 (71.1) 8 (21.1) 3 (7.9) 15 (39.5) 23 (60.5) 10 22 2 4
(26. 3) (57.9) (5.3) (10.5)
19 (50.0) 19 (50.0) 9 (23.7) 10 (26.3) 19 (50.0) 3 (7.9) 35 (92.1) 19 (50.0)
Securing system, n (%) None Sterile strips Non-sterile strips
33 (23.6) 102 (72.9) 5 (3.6)
5 (13.2) 31 (81.6) 2 (5.3)
Extension tubing, n (%) Bioconnector, n (%) Infusion pump, n (%)
13 (9.3) 131 (93.6) 27 (19.3)
7 (18.4) 36 (94.7) 2 (5.3)
to 20 days and treated with anticoagulants, had a greater incidence of phlebitis. Similarly, at bivariant level, the features of the PVC features and its care that formed a risk profile for phlebitis were: a longer PVC indwell time---it would have been inserted in the morning---a larger bore, using a transparent dressing and any other securing means, using an extension tube as an accessory and less use of an infusion pump (Table 2). However, when a controlled multivariant analysis was performed to monitor the effect of all the variables (Table 3), the only features the PVC care associated significantly with phlebitis were: insertion of a 22 or 24 gauge PVC, which reduced the risk (OR: 0.23; CI95%: 0.07---0.80), and the PVC having an
5
a
OR adjusted by the table variables and by: sex, age (<65 years, 65---79 years or ≥80 years), duration of hospital stay (<10 days, 10---19 days or ≥20 days), arterial hypertension, diabetes, neoplasm, weight loss, fever, oral anticoagulants, intravenous antibiotics (none, non irritant or irritant), intravenous chemotherapy agents (none, non aggressive or aggressive), analgesia and blood/blood products.
extension tube, which increased the risk (OR: 4.82; CI95%: 1.05---22.13).
Discussion According to the results of our prospective study, the incidence rate of phlebitis in a Spanish reference hospital is greater than the theoretically acceptable level. Using a
+Model
ARTICLE IN PRESS
6 wider definition of phlebitis (all grade I or higher), we found that a larger PVC bore and the use of extension tubing were the only variables associated with the insertion of the PVC or with its care which significantly increased the risk of phlebitis. There is enormous disparity in the rates of phlebitis in the scientific literature, which appears to be more associated with the lack of consensus on a definition that would enable a clear diagnosis than to differences in relation to the care of PVC.18 On the one hand, the incidence of phlebitis that we found using the strict criteria was practically the same as that reported in the Iberoamerican Study on Adverse Events in Latin-American hospitals19 (5.7%) and only slightly higher than that found by Ferrete-Morales et al.2 (4.8%) prior to the implementation of a specific PVC action protocol. However, on the other hand, the incidence rate found by the COSMOS study in a third level Spanish hospital was very much higher than ours, varying between 31 and 45 cases of phlebitis per 100 PVC days.20 In any case, in most of the studies performed in Anglo-Saxon countries, phlebitis is diagnosed the minute there is pain at the insertion site (grade I), i.e., using what we call the wider criterion. In these studies, the rates of phlebitis tend to vary between 20% and 30%,21,22 our estimate also found this range. In the United States, the phlebitis rate acceptable to the Infusion Nursing Society is 5% or less in all populations,23 therefore we could consider that the incidence in Asturias’ Hospital Universitario Central is only moderately acceptable using the strict criterion, and improvable if we use the wider criterion. An examination of the incidence rate of phlebitis using the removal criterion---which constituted a measurement of the staff nurses’ clinical judgement---and comparing it with what is considered the gold standard, revealed that staff nurses have high sensitivity in diagnosing phlebitis, which results in a poor positive predictive value of their evaluations, in other words, only half of the cases where the nurses had decided to remove the PVC actually had phlebitis. In the same regard, Marsh et al.18 found a low rate of agreement using different scales for evaluating phlebitis, this result raises the question of the need for agreed valid and reliable measurements of phlebitis. Furthermore, it is well known that early removal of a PVC results in major financial losses, discomfort for patients and hinders nurses undertaking other tasks.24 With regard to nursing care that might affect the risk of phlebitis, the multivariant analysis revealed that, irrespective of the other factors, only 2 nursing decisions regarding the insertion and care of a PVC were clinically relevant: the bore and the use of tube extensions. Amongst the mechanical factors, the PVC bore is one of the most studied. In general, there is relative consensus in the scientific literature with regard to the relationship between a larger PVC bore and a greater incidence of phlebitis, possibly due to the physical trauma caused to the vein intima.25---28 For this reason, it would seem reasonable to choose the lowest possible PVC bore, although obviously as the international guidelines recommend, what it is to be used for, the risk of other complications and patient comfort should be taken into account.14 However, it is still quite routine to choose
L. Arias-Fernández et al. the largest bore that the patient can tolerate, with the futile objective of avoiding obstructions that might cause a delay in nursing tasks. Finally, the effects of extension tubes have been broadly debated, with no clear recommendations on the subject. In Spain, studies that have attempted to investigate the association between PVC extension tubes and the risk of phlebitis attribute a protective effect to their use, but we consider that these outcomes might be outdated because of the modifications that have been made in PVC materials and accessories.29 For example, the study by Rivas et al.30 established that placing extension tubing between the PVC and the 3-way key reduced the risk of phlebitis, perhaps because this reduces the traction exercised by the weight of the key on the tip of the PVC. However, 3-way keys are used so seldom nowadays that our outcomes are not comparable. In fact, none of the patients in our series had a 3-way key. More recently in a doctoral thesis, Nu˜ nezCrespo31 found no significant relationship either between the use of extension tubing and the incidence of phlebitis, but the lack of association might be due to the fact that the study lacked the statistical power necessary to make this contrast, because, in fact, in that series, only the patients with extension tubing presented phlebitis. The lack of statistical power to correctly assess the risk factors is a problem that cuts across most phlebitis studies.32 Finally, according to the results of the COSMOS study, closed systems---i.e., with extension tubing incorporated into the PVC---reduce the risk of phlebitis and costs compared to accessory extension tubing (as occurred in our study).20 Therefore, it appears reasonable to avoid the use of accessory extension tubing. In any case, we were not able to establish whether it is the extension tubing itself or accessing it while failing to respect the recommended asepsis that constitutes the risk factor. The latter is unlikely because other similar accessories would then also have been demonstrated as risk factors (e.g. bioconnectors). Our study has certain limitations. Firstly, despite having an appropriate sample size to establish the incidence of phlebitis with a margin of error of less than 3%, the sample could not be sufficient to study the effect of some potential risk factors. Secondly, the selection of the participants could also have been a source of error, as the sample was not random. Nonetheless, we tried to collect a sample from the 2 large hospital areas (a medical floor and a surgical floor) and from other more specific areas (haematology and oncology). Finally, some risk of residual confusion cannot be ruled out since there might be other important variables which have not been gathered in this study, and therefore not checked in the analysis. In conclusion, the lack of consensus on the diagnosis of phlebitis makes it difficult to assess this complication. Staff nurses tend to make a diagnosis of phlebitis and to remove the PVC before it is clinically advisable, which can reduce the real incidence of phlebitis but it can also increase patient discomfort and the costs for the system. According to the results of studying the risk factors, in order to minimise the incidence of phlebitis it would be advisable to choose the smallest bore of PVC possible and avoid the use of accessory extension tubes.
+Model
ARTICLE IN PRESS
Incidence and risk factors of phlebitis associated to peripheral intravenous catheters
Conflict of interests The authors have no conflict of interests to declare.
14.
Acknowledgements The authors would like to thank the supervisors of the hospital units included in the study for their collaboration: Eugenia García, M. Antonia Duarte, Susana Campos and Carmen Vena. We would also like to thank Julián Díaz for his participation and advice in piloting the data collection instrument. This research was conducted in the context of the Flebitis-Zero project (http://flebitiszero. com/app/).
15.
16. 17.
References 1. Alianza Mundial para la Seguridad del Paciente. La Investigación en Seguridad del Paciente [Internet]. París: Organización Mundial de la Salud; 2008. Available in: http://www.who.int/patientsafety/information centre/ documents/ps research brochure es.pdf [accessed 28.11.15]. 2. Ferrete-Morales C, Vázquez-Pérez MA, Sánchez-Berna M, Gilabert-Cerro I, Corzo-Delgado JE, Pineda-Vergara JA, et al. Incidencia de flebitis secundaria por catéter venoso de acceso periférico e impacto de un protocolo de manejo. Enferm Clin. 2010;20:3---9. 3. Pasalioglu KB, Kaya H. Catheter indwell time and phlebitis development during peripheral intravenous catheter administration. Pak J Med Sci. 2014;30:725---30. 4. Nassaji-Zavareh M, Ghorbani R. Peripheral intravenous catheter related phlebitis and related risk factors. Singap Med J. 2007;48:733---6. 5. Uslusoy E, Mete S. Predisposing factors to phlebitis in patients with peripheral intravenous catheters: a descriptive study. J Am Acad Nurs Pract. 2008;20:172---80. 6. Washington GT, Barret R. Peripheral phlebitis: a pointprevalence study. J Infus Nurs. 2012;35:252---8. 7. Kaur P, Thakur R, Kaur S, Bhalla A. Assessment of risk factors of phlebitis amongst intravenous cannulated patients. Nurs Midwifery Res J. 2011;7:106---14. 8. Gallant P, Schultz AA. Evaluation of a Visual Infusion Scale for determining appropriate discontinuation of peripheral intravenous catheters. J Infus Nurs. 2006;29: 338---45. 9. Hadaway L. Short peripheral intravenous catheters and infections. J Infus Nurs. 2012;35:230---40. 10. Lee WL, Chen HL, Tsai TY, Lai IC, Chang WC, Huang CH, 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---6. 11. Rickard CM, Webster J, Wallis MC, Marsh N, McGrail MR, French V, et al. Routine versus clinically indicated replacement of peripheral intravenous catheters: a randomised controlled equivalence trial. Lancet. 2012;380: 1066---74. 12. O’Grady NP, Alexander M, Burns A, Dellinger EP, Garland J, Heard SO, et al. Guidelines for the Prevention of Intravascular Catheter-Related Infections [Internet]. Atlanta: Centers for Disease Control and Prevention (CDC); 2011. Available in: http://www.cdc.gov/hicpac/pdf/guidelines/bsi-guidelines2011.pdf [accessed 14.10.15]. 13. Webster J, Osborne S, Rickard CM, New K. Clinically-indicated replacement versus routine replacement of peripheral
18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
28.
29.
7
venous catheters. Cochrane Database Syst Rev. 2015;8: CD007798. Loveday HP, Wilson JA, Pratt RJ, Golsorkhi M, Tingle A, Bak A, et al. epic3: National evidence-based guidelines for preventing healthcare-associated infections in NHS hospitals in England. J Hosp Infect. 2014;86 Suppl. 1: S1---70. González E, Argüelles AR, Martínez B. Comisión de Cuidados de Enfermería. Protocolo para la inserción, mantenimiento y retirada del catéter venoso periférico [Internet]. Oviedo: Hospital Universitario Central de Asturias; 2013. Available in: http://www.hca.es/huca/web/enfermeria/html/f archivos/ Cateter%20venoso%20periferico.pdf [accessed 20.06.16]. Jackson A. Infection control-a battle in vein: infusion phlebitis. Nurs Times. 1998;94:68---71. Martínez-Ortega C, Suárez Mier B (coordinadoras). Flebitis Zero: diagnóstico de la flebitis [Internet]. Oviedo: Hospital Universitario Central de Asturias; 2015. Available in: http://flebitiszero.com/app/formacion/DiagnosticoFlebitis. html [accessed 20.06.16]. Marsh N, Mihala G, Ray-Barruel G, Webster J, Wallis MC, Rickard CM. Inter-rater agreement on PIVC-associated phlebitis signs, symptoms and scales. J Eval Clin Pract. 2015;21: 893---9. Aranaz JM, Aibar C, Larizgoitia I, Gonseth J, Colomer C, Terol E, et al. Estudio IBEAS. Prevalencia de efectos adversos en hospitales de Latinoamérica [Internet]. Madrid: Ministerio de Sanidad y Política Social; 2010. Available in: http://www.seguridaddelpaciente.es/resources/contenidos/ castellano/2009/INFORME IBEAS.pdf [accessed 03.05.15]. González López JL, Arribi Vilela A, Fernández del Palacio E, Olivares Corral J, Benedicto Martí C, Herrera Portal P. Indwell times, complications and costs of open vs closed safety peripheral intravenous catheters: a randomized study. J Hosp Infect. 2014;86:117---26. Davies J, Cotton J, Wood J, Burke D, McMahon M. Intravenous nutrition----peripheral or central administration? Br J Surg. 2003;90 Suppl. 1:86---91. Grune F, Schrappe M, Basten J, Wenchel HM, Tual E, Stutzer H. Phlebitis rate and time kinetics of short peripheral intravenous catheters. Infection. 2004;32:30---2. Infusion Nurses Society. Infusion nursing ---- standards of practice. J Infus Nurs. 2011;34 Suppl. 1:1---106. Hasselberg D, Ivarsson B, Andersson R, Tingstedt B. The handling of peripheral venous catheters----from noncompliance to evidence-based needs. J Clin Nurs. 2010;19: 3358---63. Wallis MC, McGrail M, Webster J, Marsh N, Gowardman J, Playford EG, 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---8. Saini R, Agnihotri M, Gupta A, Walia I. Epidemiology of infiltration and phlebitis. Nurs Midwifery Res J. 2011;7: 22---33. Tagalakis V, Kahn SR, Libman M, Blostein M. The epidemiology of peripheral vein infusion thrombophlebitis: a critical review. Am J Med. 2002;113:146---51. Cicolini G, Manzoli L, Simonetti V, Flacco ME, Comparcini D, Capasso L, et al. Phlebitis risk varies by peripheral venous catheter site and increases after 96 h: a large multi-centre prospective study. J Adv Nurs. 2014;70: 2539---49. López Rodríguez L, Seda Diestro J, Garrido Díaz-Malaguilla IM. Llave de tres vías con alargadera en complicaciones asociadas a la vía intravenosa. Enferm Clin. 1998;8: 104---9.
+Model
ARTICLE IN PRESS
8 30. Rivas JS, Artes JL, Arjona FJ, Carmona D, Soriano FJ, Mejías FM, et al. Flebitis traumáticas. Cómo disminuir su incidencia. Rev Enferm. 2004;27:602---6. 31. Nú˜ nez-Crespo FJ. Efectos adversos hospitalarios en dispositivos venosos periféricos: estudio de validez diagnóstica [tesis]. Madrid: Universidad Rey Juan Carlos; 2014.
L. Arias-Fernández et al. 32. Webster J, McGrail M, Marsh N, Wallis M, Ray-Barruel G, Rickard CM. Post infusion phlebitis: incidence and risk factors. Nurs Res Pract. 2015;2015:691934.