Abstract
Background:
In a previous trial, in-line filtration significantly prevented postoperative phlebitis associated with short peripheral venous cannulation. This study aims to describe the cost-effectiveness of in-line filtration in reducing phlebitis and examine patients’ perception of in-hospital vascular access management with and without in-line filtration.
Methods:
We analysed costs associated with in-line filtration: these data were prospectively recorded during the previous trial. Furthermore, we performed a follow-up for all the 268 patients enrolled in this trial. Among these, 213 patients responded and completed 6 months after hospital discharge questionnaires evaluating the perception of and satisfaction with the management of their vascular access.
Results:
In-line filtration group required 95.60€ more than the no-filtration group (a mean of € 0.71/patient). In terms of satisfaction with the perioperative management of their short peripheral venous cannulation, 110 (82%) and 103 (76.9%) patients, respectively, for in-line filtration and control group, completed this survey. Within in-line filtration group, 97.3% of patients were satisfied/strongly satisfied; if compared with previous experiences on short peripheral venous cannulation, 11% of them recognised in-line filtration as a relevant causative factor in determining their satisfaction. Among patients within the control group, 93.2% were satisfied/strongly satisfied, although up to 30% of them had experienced postoperative phlebitis. At the qualitative interview, they recognised no difference than previous experiences on short peripheral venous cannulation, and mentioned postoperative phlebitis as a common event that ‘normally occurs’ during a hospital stay.
Conclusion:
In-line filtration is cost-effective in preventing postoperative phlebitis, and it seems to contribute to increasing patient satisfaction and reducing short peripheral venous cannulation–related discomfort
Introduction
Phlebitis is a frequent complication associated with peripheral intravenous (IV) infusions, occurring in up to 96% of all hospitalised patients.1–5 The clinical features associated with this venous inflammation are usually burdensome for the patient and encompass pain, tenderness, swelling, induration, warmth and palpable cord-like veins. Phlebitis often requires the removal of the venous catheter and its replacement in a different site, causing further patient distress, time consumption for both nurses and physicians and increasing healthcare costs. Even more, repeated episodes of phlebitis and vascular access replacements progressively reduce the site for venous cannulation and might require more invasive procedures, such as central venous cannulation or deep vein access via ultrasound guidance. Micro- and nano-particles, as well as air bubbles unintentionally infused through the venous line, have been all recognised as significant risk factors for infusion-related phlebitis.1,6,7 In-line filtration effectively removes particles, air bubbles and endotoxin in clinical practice, and thus it is potentially able to reduce phlebitis associated with short peripheral venous cannulation (SPVC).8–11 Nevertheless, the costs related with in-line filters might burden their clinical use.
The effect of in-line filtration on postoperative phlebitis associated with SPVC was assessed in a previous randomised, controlled, clinical trial on hospitalised patients scheduled for major abdominal surgery and undergoing SPVC. 12 In this trial, in-line filtration was demonstrated as being able to independently reduce the incidence and severity of postoperative phlebitis, as well as prolonging venous cannula lifespan. 12 These protective effects on postoperative phlebitis, as well as the prevention of repetitive venous cannulation, might reduce the patients’ discomfort and the healthcare costs associated with the peripheral vascular access management.
This study aims to describe the cost-effectiveness of in-line filtration in reducing postoperative phlebitis associated with SPVC and evaluate differences in patients’ perception of in-hospital vascular access management with and without in-line filtration.
Methods
A previous randomised, controlled, clinical trial, was performed at the Department of Anaesthesia and Intensive Care of the Azienda Ospedaliera – Universitaria Careggi in Florence, Italy, on the effect of in-line filtration. In this trial, in-line filtration significantly reduced the incidence and severity of postoperative phlebitis associated with SPVC and prolonged the venous cannulae lifespan in a cohort of hospitalised patients scheduled for major abdominal surgery and undergoing SPVC. 12 Briefly, for patients randomised to in-line filtration, in-line filters (Pall, Dreieich, Germany) were used during anaesthesia and the following 96 postoperative hours. All drugs administered intraoperatively and postoperatively in patients enrolled in the in-line filter group had in-line filtration before endovascular infusion into the patients. Use and schematic placement of in-line filters within the infusion set were described in the previous article. 12
A cost analysis was performed on the data gathered during that trial to evaluate the economic impact of in-line filtration on the routine management of patients’ vascular access. In particular, costs for venous cannulation, vascular access dressing, infusion set, in-line filters and medical/nursing work time were prospectively recorded and compared between the filter and no-filter group.
Furthermore, to identify the qualitative differences in patients’ perception on in-hospital vascular access management between the filter and no-filter group, a follow-up of all the 268 patients enrolled in the previous trial was performed 6 months after hospital discharge. Their responses were analysed using content analysis to identify similarities and differences within and between the groups. The questionnaire contained quantitative variables exploring the patient’s accordance with specific statements (dichotomous answers: agree/disagree) and the patient’s satisfaction regarding well-defined aspects of the vascular access management (categorised answers: strongly-unsatisfied/un-satisfied/neutral/satisfied/strongly satisfied). Furthermore, qualitative variables, mainly represented by questions with open answers, were introduced into the questionnaire and the main domains explored through qualitative analysis. In the qualitative analysis, three domains were primarily identified in the patients’ descriptions:
Patient’s discomfort for daily activities or functional limitation during the 96 h of the study observation.
General patient’s satisfaction in vascular access management during the 96 h of the study observation.
Patients’ perspective on comparison with previous in-hospital experiences requiring SPVC.
This study was approved by the Local Review Board (Comitato Etico di Area Vasta, CEAV IN17/0000015), and written informed consent was obtained from all subjects participating in this trial (clinicaltrials.gov NCT03193827, Principal investigator: Dr Gianluca Villa, Date of registration: 18 June 2017).
Data analysis
We analysed distribution of quantitative data through the Shapiro–Wilk test. All quantitative data are presented as medians and interquartile range (IQR) or mean ± standard deviation (SD), where appropriate. Qualitative variables are expressed as percentages.
Bivariate association analyses have been assessed through t-student or Fisher’s exact test depending on the variable analysed. Associations are described as Odds Ratio (OR), 95% confident interval (95%CI) and p-value; p < 0.05 has been considered statistically significant.
Qualitative analysis has been conducted considering the free answers that the patients gave during the telephone interviews. In particular, the domains that most frequently occurred during these answers were identified, and the patients’ perception reported.
Results
During the 96 h of observation, a mean of €10.89 (95%CI €8.30–€13.48) was spent for each patient in the filter group, while a mean of €9.92 (95%CI €4.44–€15.41) for each patient in the no-filter group. Specific items considered for cost analysis are reported in Table 1. The cumulative cost analysis has shown that the vascular access management in the in-line filtration group required €95.60 more than in the no-filtration group (€0.71/patient, see Table 1).
Specific costs recorded among filter and no-filter group.
A total of 213 of the 268 patients were available and completed questionnaires evaluating the perception of and satisfaction with the management of their vascular access. The selection process is described in Figure 1.
The selection process for the qualitative/quantitative questionnaire.
The clinical characteristics of patients considered for this survey are described in Table 2.
Patients’ clinical characteristics.
BMI: body mass index.
Three patients (2.7%) in the filter group and 7 (6.8%) in the no-filter group were strongly unsatisfied or unsatisfied or neutral regarding the management of their vascular access; while patients satisfied or strongly-satisfied on the perioperative management of their short peripheral venous cannula were 107 (97.3%) and 96 (93.2%), respectively, for filter group and no-filter group (p = 0.20, OR 2.59, 95%CI 0.57–15.95).
Four patients (3.6%) in filter group and 8 patients (7.8%) in no-filter group evaluated their discomfort during daily activities as excessive (p = 0.24, OR 0.45, 95%CI 0.10–1.74). Interestingly, 96 patients (93.2%) in the no-filter group were satisfied/strongly satisfied on the perioperative management of their short peripheral venous cannula, although up to 30% of them experienced postoperative phlebitis. A total of 88 patients (80%) in the filter group and 72 patients (69.9%) in the no-filter group underwent previous hospitalisation requiring a venous cannula for more than 24 h. These patients compared the experience they had during the trial with these previous experiences. Interestingly, 9 patients (11%) in the filter group remarked improvement in vascular access management compared to the previous experiences, while no patient in the no-filter group mentioned it (Table 3).
Clinical characteristics for strongly satisfied/satisfied (SS/S) and strongly unsatisfied/unsatisfied/neutral (SU/U/N) patients, both in filter and no-filter group.
SPVC: short peripheral venous cannulation.
No differences in the prevalence of post-phlebitis symptoms have been shown between the two groups of patients (Table 4).
Post-phlebitis symptoms in filter and no-filter group.
In the qualitative analysis, three domains were primarily identified in the patients’ descriptions:
4. Patient’s discomfort for daily activities or functional limitation during the 96 h of the study observation. Two of the three ‘strongly unsatisfied/unsatisfied/neutral’ patients in the filter group described their experience as uncomfortable for the cannula hand location which, ‘precluded free movement’ during the hospitalisation. The third patient described that, ‘the presence of the filter reduced the flow of the drugs into the venous line [mainly driven by gravity; the peristaltic pump was not used], prolonging the time required for infusion’.
5. General patient’s satisfaction in vascular access management during the 96 h of the study observation. Patients strongly satisfied/satisfied in the filtration group would like the ‘introduction of in-line filtration in vascular access management in case of future hospitalisations’. In particular, one of patient demanded in-line filtration in case of a further admission in our centre reporting that, ‘that is the only way to prevent phlebitis that I always experience during the hospital length of stay’.
6. Patients’ perspective on comparison with previous in-hospital experiences requiring SPVC. Patients with previous SPVC experiences remarked ‘great improvement in vascular access management’ during in-line filtration. In particular, most of the strongly-satisfied/satisfied patients in the filtration group declared that ‘it was the first hospitalisation without phlebitis in my life’. Interestingly, the qualitative interview of the satisfied/strongly-satisfied patients in the no-filter group who had phlebitis postoperatively during the 96 h of the trial observation, reveals fascinating aspects related with SPVC: (1) these patients recognised ‘no differences than previous experiences on SPVC’. (2) These patients mentioned postoperative phlebitis as ‘a fact’ or as ‘normal consequences, not even a complication’ that ‘normally occurs’ during SPVC and hospital stay. (3) Although these patients experienced postoperative phlebitis and more than one short cannula replacement during hospitalisation, they ‘didn’t expect any differences in that [referred to the hospitalisation in which the enrolment for trial occurred] or in future hospitalisation regarding the vascular access management’.
Discussion
The rate of venous postoperative phlebitis associated with SPVC and cannula malfunction is high in hospitalised patients undergoing surgery; up to 69% of these devices fail before treatment is complete and need to be replaced. 13 This phenomenon often leads to interruptions in pharmacological therapies, needs for venous cannula replacement, increases in-hospital length of stay, morbidity and mortality, wastes in medical/nursing time, and higher costs.
Because of the removal of particles (e.g. plastic/glass particulate, microscopic cellulose fibres, fungal spores, drugs aggregates, micro and macro rubber particles), air bubbles, and the prevention of recurring handling of the infusion lines, in-line filtration has been proposed as a potential tool to reduce phlebitis.14,15 In a previous randomised controlled trial on 268 surgical patients undergoing SPVC, we demonstrated a significant reduction on the incidence of postoperative phlebitis (defined through VIP score) correlated with in-line filtration (2.2% vs 26.9%, difference (95%CI) of 25% (12%–36%) OR 0.05 (0.01–0.15), p < 0.001). Furthermore, patients in the no-filter group had higher VIP scores in the 24–96 h postoperatively and the dwell time of the short cannulae in the in-line filter group was longer than in the no-filter group. Interestingly, filtration was used in this trial during anaesthesia and the following 96 postoperative hours. Especially (not only) for colorectal elective procedures, there is growing evidence internationally to support Enhanced Recovery After Surgery (ERAS) pathways. Although the impact of pathway compliance and the relative importance of individual components remains poorly clear, a significant reduction in IV administration of drugs and fluids could impact the need of postoperative use and duration of SPVC, probably less than the 96 h of this trial.
The cost analysis performed in this study shows that this significant reduction in postoperative phlebitis associated with SPVC was associated with only a very limited increase in the healthcare cost per single patient (less than 1 € for each patient). This is an important finding, as costs related to the filters have been cited as a limit to their wide-spread use. 16 In our study, the costs associated with in-line filters in the study group were offset by the cost associated with the change of infusions’ lines and nurses/physician hourly costs in the control group. This finding is in accordance with that observed by Van Den Hoogen et al. 17 in a randomised clinical trial on paediatric critical care patients. They demonstrated that the costs associated with in-line filtration were similar to that observed in the control group. In particular, the prices for in-line filter and the replacement of the IV lines once every 96 in the study group were comparable with those for the daily change of all IV-administration sets in the group without filters. Furthermore, with a mean nursing time of 10 ± 5 and 14 ± 7 min, respectively, in the study and control group (p < 0.0001), they further demonstrated that a significantly longer time was necessary for changing the IV-administration sets in the non-filter group. They concluded that the increased costs for in-line filters were compensated by the more extended use of IV administration sets and by the reduced nurses staffing.
Similarly, in our study, the total cost for in-line filters (957€) in the filter group was counterbalanced with increased expenditure for infusion sets change (766.7€ vs 266.7€) and nursing/medical time of work in no-filter group (542.5€ vs 189.7€). These costs are entirely explainable with the reduced dwell time of the short cannula in the no-filter group, leading to its removal and the subsequent replacement.
Although this cost analysis is strongly influenced by local and national variables, such as the costs for disposable materials applicable to our hospital or staff salary, in our study in-line filtration seems to be cost-effective in reducing postoperative phlebitis among surgical patients.
These results are of further interest if evaluated in unison with the data deriving from the patients’ perception on the quality of in-hospital management of peripheral short cannula.
In particular, the quantitative/qualitative survey performed among patients enrolled in the previous trial show that in-line filtration modifies the satisfaction perceived for the vascular access management. A statistical difference between the filter and no-filter group was not observed on the number of satisfied/strongly satisfied patients (97.3% vs 93.2%, p = 0.20). Nevertheless, most patients in the filter group who previously underwent SPVC remarked on a significant improvement in vascular access management and identified in-line filtration as an innovation they would like to have for future hospitalisations. Interestingly, most of strongly unsatisfied/unsatisfied/neutral patients in the filtration group correlated their dissatisfaction to the management and location of their peripheral vascular access, not to the presence of the in-line filter. In fact, they justified their reduced satisfaction as a result of a lowered freedom in arm movement related with an unconventional short cannula location (vein in the dorsum of the hand, for example). In these terms, it is essential to remark that in-line filtration alone can neither reduce postoperative phlebitis nor improve patients’ satisfaction on SPVC; only that if it is applied in conjunction with all other clinical guidelines on cannulation and management of peripheral short cannula, these filters might enhance the in-hospital clinical practice.
More interestingly, satisfied/strongly satisfied patients in the no-filter group have defined phlebitis as ‘normal occurrence’ during hospitalisation, almost to define phlebitis as an inevitable event. Therefore, there are no patients in the no-filter group who have found differences or improvement in vascular access management compared with the previous hospitalisation. This high expectation of postoperative phlebitis in the no-filter group can explain the high incidence of satisfied/strongly satisfied patients in this group. To the best of our knowledge, this is the first study aimed to evaluate the patients’ perspective on in-line filtration, both qualitatively and quantitatively. No comparison with previous results is thus available.
We acknowledge that patients’ perceptions and satisfaction with the management of their vascular access is a difficult issue to be explored, even in a randomised controlled trial (RCT). Indeed, psychological conditions of patients receiving a filtering device for the completion of intra-postoperative treatment might be different from that of patients experiencing the usual approach. Furthermore, each patient in the study group knew that he was being treated with filters and knew their potential beneficial effects. Finally, too many factors affect the patients’ satisfaction when they receive a short peripheral venous cannula, such as the site of insertion and the type of securement/dressing. All these issue might be recognised as potential drawbacks of the study with potential consequences on the reliability of perceptions measured by the qualitative questionnaires.
Conclusion
In-line filtration reduces the incidence and the severity of postoperative phlebitis associated with SPVC without increased healthcare costs. Although several confounding factors might limit qualitative results in this study, in-line filtration seems to have a potential role in improving patient satisfaction for the in-hospital short peripheral cannula management, and in reducing SPVC-related discomfort.
Footnotes
Author contributions
G.V. substantially helped in conception and design, acquisition of data and analysis and interpretation of data. R.G. helped in conception and design and acquisition of data. T.A. helped to draft the article or review it critically for important intellectual content. L.T. helped in analysis and interpretation of data. C.C. helped to draft the article or review it critically for important intellectual content. F.P. was accountable for all aspects of the work thereby ensuring that questions related to the accuracy or integrity of any part of the work were appropriately investigated and resolved. A.R.D.G. gave final approval of the version to be published. S.R. helped to draft the article or review it critically for important intellectual content;
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: We have the following interests: G.V. received from Baxter support for travel expenses, hotel accommodations and registration to meetings. R.G. received from Pall International support for travel expenses, hotel accommodations and registration to meetings. C.C. received honoraria for lectures by Orion Pharma; a grant for consultancy by Astellas; support for meetings (travels, hotel accommodations and/or registration) by BBraun, Astellas, MSD, Pfizer, Pall International, Baxter and Orion Pharma. F.P. received from MSD, Orion Pharma, Vygon, BBraun, Medcomp, Bard and Teleflex, support for travel expenses, hotel accommodations and registration to meetings. From Bard received honoraria for lectures. A.R.D.G. received research grants from MSD Italia, Baxter, Pall international. S.R. received from Baxter, Orion Pharma and Vygon honoraria for lectures, from ICU Medical, MSD and Medtronic grants for consultancy, from Baxter, BBraun, Pall International and Vygon support for travel expenses, hotel accommodations and registration to meetings. S.R. received from Baxter, Orion Pharma and Vygon honoraria for lectures, from ICU Medical, MSD and Medtronic grants for consultancy, from Baxter, BBraun, Pall International and Vygon support for travel expenses, hotel accommodations and registration to meetings. All other authors have no interests to declare. There are no patents, products in development or marketed products to declare. This does not alter our adherence to all the JVA policies on sharing data and materials.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Pall International and ‘Philip and Irene Toll Gage Foundation’. These funders have provided a grant to the Department of Health Science of the University of Florence to economically support the feasibility, management, coordination, statistical analysis and recruitment of investigators for this study. Furthermore, the National Heart Lung and Blood Institute (NHLBI Grant No.: T32 HL134625) has economically supported this study. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.