Perceptions of risk factors for phlebitis among Malaysian nurses

Background

IV therapy is an integral part of professional nursing practice that aims to infuse fluids, administer IV medication, transfuse blood products, and deliver nutrients to patients (Ho and Cheung, 2012). Tan et al (2012) found that 35.2% of the 428 adult patients with a peripheral cannula in surgical and medical wards had thrombophlebitis in a tertiary hospital in the central region of Malaysia. An observational study by Azlina (2018) found that the incidence of phlebitis was 36.1% in one public hospital in the east coast of Malaysia. The authors of the present study found these relatively high incidences of phlebitis in these studies in Malaysian hospitals worrying, which prompted them to undertake this research. Studies undertaken by Wilfong et al (2011), Hadaway (2012) and Arbaee and Mohd Ghazali (2013) reported that nurses' knowledge and early recognition of risk factors for the development of phlebitis could reduce this complication. Nurses have a significant role in preventing phlebitis. Early recognition will enable prompt intervention, minimise disruption to treatment and reduce the incidence of phlebitis.

Aim

This study aimed to investigate nurses' perceptions of the risk factors for phlebitis in a tertiary teaching hospital in north-east Peninsular Malaysia. Its ultimate objective was to help healthcare policymakers decrease the prevalence and incidence of phlebitis in patients receiving IV therapy by providing in-service nurse education on the prevention of phlebitis.

Design

A cross-sectional study was carried out in Hospital Universiti Sains Malaysia (Hospital USM) located in east coast Peninsular Malaysia.

Participants

The study population comprised registered nurses based in medical, surgical, multidisciplinary (a medical-surgical ward), oncology, and orthopaedic wards; and with at least 6 months of clinical experience. According to Aqtash et al (2017), nurses with 6 months' or more experience of clinical practice were considered competent in providing direct patient care. Sample size was calculated based on Ho et al's (2016) study using the single mean formula. Assuming the precision value of 0.5, and a 95% confidence interval, the minimum sample size was 181 nurses. After considering a likely dropout rate of 10%, a total of 199 nurses were recruited for this study and were randomly selected using Excel from the list of nurses obtained from the nursing unit. This sampling offered an equal chance of selection for everyone within the population group. The researcher (CXY) then personally contacted the participants.

Data collection

Data collection was carried out from February 2018 to March 2018. The research was explained verbally by the researcher, a final year undergraduate nursing student, not known to the nurses. Willing participants completed a written consent. The participants were given a copy of the questionnaire, which took approximately 10 to 15 minutes to complete and was then collected by the researcher. The consent forms and data obtained were kept by the researcher's supervisor in a lockable room. During the study, the data were stored on the researcher's computer, access to which was restricted by a code only known to the researcher. All data will be kept for 2 years after publication of the research and then destroyed. The questionnaire was self-administered and contained two sections, A and B. Section A was designed to collect the nurses' demographic data (gender, age, years of clinical experience, the highest level of nursing education, and clinical area). Section B contained 20 statements and was used and modified with permission from its authors (Li et al, 2016). Pilot testing of the measurement instrument was carried out on 20 nurses from an otorhinolaryngology ward, who were not involved in the main study, to assess the feasibility and reliability of the questionnaire. The questionnaire was originally designed for use with Chinese nurses and was only slightly modified to fit the Malaysian nurse context. The modified questionnaire contained 20 items after the pilot test (compared with 14 items in the original). The modified questionnaire included wording believed to be relevant to nurses' perceptions of risk factors for phlebitis among their patients, such as the improper cannula size and location, pharmacological properties and osmolality, flow rate, presence of associated diseases in patients, duration of IV therapy and poor aseptic techniques. The scoring procedure was also modified. In this survey, only positive statements were used as the combination of positive and negative statements can affect the internal consistency of the scales and negative items tend to be more intercorrelated (Solís Salazar, 2015). A Cronbach alpha of 0.84 was obtained, indicating a good level of item reliability (Denzin and Lincoln, 2017).

Participants were required to respond to statements with ‘Yes’, ‘No’ or ‘Unsure’. The ‘Unsure’ option was designed to avoid unnecessary Yes/No guessing. One point was awarded for each correct response and zero was given to each wrong or unsure response. The total points obtained were converted into a percentage mark from the raw mark. The level of perception was interpreted based on the percentage marks obtained, which were classified in three categories:

The questionnaire was distributed to each eligible nurse either at the end of their shift, during their break or at a time convenient for them. Participants were asked not to use any resources or confer with colleagues while completing the questionnaire. The researcher collected the completed questionnaires.

Ethical considerations

This study was approved by the Ethics Review Committee of Universiti Sains Malaysia and the Director of Hospital Universiti Sains Malaysia, Malaysia. The study was performed based on the principles of the Declaration of Helsinki. All participants provided written informed consent. All participation was voluntary. No information was collected regarding participants' identity and completed consent forms and responses were collected separately to ensure confidentiality and anonymity.

Data analysis

The data were entered into the Statistical Package for Social Science (SPSS) software version 25.0. Following checks for accuracy of data entry, descriptive statistics were used to identify frequencies/valid percentages and measures of central tendency and dispersion, for example, standard deviation, where appropriate. Nurses' demographic factors were described as frequency and percentage. However, nurses' age and years of clinical experience were presented in mean and standard deviation formats because the raw data were numerical. Frequency and percentage were also used to describe the statistics for nurses' perceptions of risk factors for phlebitis. Fisher's exact test was used to analyse the association between selected demographic factors and perceptions of risk factors for phlebitis. Values of P≤0.05 were considered significant.

Results

Nurses' perceptions of risk factors for phlebitis

More than half of the studied population (56.8%) had good perceived phlebitis risk factors knowledge about acidity, osmolality, and concentration of infusion drugs, and what this meant for the risk of phlebitis. A high percentage understood healthcare-related risks. However, the results indicated some shortcomings in nurses' knowledge about the risks of phlebitis in some areas of patient-related risk factors, with more than 50% not recognising that replacing the cannula more frequently than every 72–96 hours (3–4 days) and using a 16G cannula increased the risk for phlebitis. Perceptions of risk factors for phlebitis among nurses and the frequency of each perception level are detailed in Table 1 and Figure 1.

Discussion

This study aimed to investigate nurses' perceptions of risk factors for phlebitis in a tertiary teaching hospital in north-east Peninsular Malaysia. Despite the published Guidelines on Safe Operating Procedure for Administration of Intravenous (Bolus) Medications for Nurses (Ministry of Health Malaysia Nursing Division, 2017), only 56.8% of participants indicated good knowledge of the perceived phlebitis risk factors such as acidity, osmolality, and concentration of infusion drugs. Normally it is expected that nurses should be knowledgeable about peripheral intravenous care. Some deficiencies in recognising cannula-related risk factors were also a cause for concern. Although some studies have suggested that the size of cannula affects the development of phlebitis, 57.3% of nurses in this study did not recognise that the use of a 16G cannula can increase the risk of phlebitis. Martinho and Rodrigues (2008) suggested that phlebitis often occurs when the size of the cannula is too big for the selected vein, resulting in unnecessary friction on the vein's internal lining, causing inflammation—known as mechanical phlebitis. Pasalioglu and Kaya (2014) suggested that phlebitis is caused by the cannula's wide and thick lumen, while Urbanetto et al (2016) found a significant relationship between the use of the larger cannula and the incidence of post-infusion phlebitis.

Also of concern is the finding that 56.3% of the nurses did not think that replacing the cannula more frequently than every 72–96 hours (3–4 days) could increase the risk of phlebitis. This is contrary to the Centers for Disease Control and Prevention (CDC) guidelines (O'Grady et al, 2011), which recommended that peripheral intravenous cannulas should not be replaced more frequently than every 72–96 hours to prevent irritation of the vein and avoid mechanical phlebitis.

A positive finding of the study was that 84.9% of the nurses understood that leaving the cannula in situ for more than 96 hours (4 days) could increase the risk for phlebitis. This not only implies good awareness of the risk of infectious phlebitis, but corresponds with Cicolini et al's (2014) findings from their multi-centre prospective field study, which revealed that the likelihood of developing phlebitis increased with increasing cannula duration, and was highest after 96 hours. Erdogan and Denat (2016) also reported that phlebitis rates reached 90–100% after 96 hours. Another positive finding was that 84.4% of the nurses believed that proper cannula securement decreases the risk for phlebitis. A systematic review carried out by Marsh et al (2015) indicated that adequate securement of the cannula can decrease phlebitis rates; O'Grady et al (2011) suggested that an inadequately secured peripheral intravenous cannula has the potential to migrate externally and fall out, which may lead to a damaged vessel wall and mechanical phlebitis.

Some 45.2% of nurses agreed that insertion of the cannula into the antecubital fossa increases the risk of phlebitis, but 28.6% disagreed and 26.1% were unsure; this was concerning. Those who disagreed thought that the larger vessels in the antecubital fossa may be less likely to become inflamed and others were unsure about the anatomical location of the antecubital fossa. Forni et al (2010), Rego Furtado (2011b) and Mestre et al (2012) have all shown higher phlebitis rates for cannulas inserted in the antecubital fossa, suggesting that excessive movements at the elbow joint cause vessel wall trauma.

In total, 88.9% of the nurses recognised drug-related risk factors for phlebitis, such as the infusion of drugs that are acidic, alkaline, have high osmolarity, or are highly concentrated. This result may be due to the nurses' familiarity with routinely administering a variety of medications and diluting them according to hospital protocols and to the instructions on the medication packages. However, during data collection, the researcher observed that, although nurses may be able to identify which drugs would normally and potentially lead to phlebitis, they may not know which specific drugs are acidic, alkaline, and have high osmolarity. This may imply a need to improve nurses' theoretical knowledge in this area. Monteiro et al (2012) stressed the importance of knowing drugs’ pH, quantity, concentration, and the appropriate dilutant to minimise the risk of phlebitis because drugs with a pH and osmolarity that are very different from those of the patient's blood increase the risk of phlebitis. The more acid a drug, the more dilution is needed to administer it in the highest volume clinically tolerated by the patient (Bitencourt et al, 2018).

Flushing of the IV giving set influences both the effectiveness and safety of intravenous therapy (Infusion Nurses Society, 2016). It maintains cannula patency by preventing internal luminal occlusion, reducing the build-up of blood and other products on the cannula's internal surface, and avoids the interaction of incompatible medications (Infusion Nurses Society, 2016). Thus, flushing is essential in maintaining an open venous cannula and reducing phlebitis (Eghbali-Babadi et al, 2015). This study showed that 88.4% of the nurses agreed that the risk of phlebitis increases if flushing is not done after, and between, each medication administration—a practice they perform in their daily routines. Wang et al (2012) concluded that 0.9% sodium chloride solution and heparin saline solution are effective and safe flushing solutions for peripheral intravenous access devices.

This study found that 78.4% of the nurses recognised that the administration of large volumes of crystalloids and colloids at high flow rates, and trauma to the vein during cannula insertion are phlebitis risk factors, which concurs with several studies. Saini et al (2011) identified administration of more than 2 litres of crystalloids such as normal saline, 5% dextrose or Ringer's lactate solution at a rate more than 50 ml/hour, colloids and medications such as antibiotics through the peripheral intravenous cannula as the most important risk factors for the development of phlebitis. Helm et al (2015) showed that mechanical trauma to the intima and vessel wall have a higher movement-related phlebitis rate, and Erdogan and Denat (2016) found a higher phlebitis rate when the cannula was inserted by health workers with less than 5 years' experience due to a higher chance of mechanical trauma.

The importance of skin hygiene and aseptic techniques during care of the cannulated site in decreasing the risk of phlebitis was acknowledged by 93% of nurses in this study, which supports other research associating poor hand hygiene or aseptic techniques with colonisation by microorganisms along the skin track or contamination of intravenous cannula (Zhang et al, 2011; Osei-Tutu et al, 2015). O'Grady et al (2011) stated that clean skin should be prepared with a >0.5% chlorhexidine with alcohol solution before central and peripheral venous cannula insertion, while, in New Zealand, the Canterbury District Health Board (2019) recommended using 2% chlorhexidine combined with 70% isopropyl alcohol as an effective skin disinfectant that minimises the risk of infectious phlebitis.

The nurses in the present study identified unclean dressings (65.3%) as an independent risk factor for phlebitis; a finding consistent with Miliani et al (2017), who reported that a soiled or unclean dressing could provide an ideal breeding ground for infection-producing microorganisms and increase the risk of infectious phlebitis. The statement that gained the greatest consensus (95%) among the nurses in this study was that daily assessment of the cannulation site for signs of phlebitis could decrease the risk for phlebitis by initiating early intervention.

In terms of patient-related risk factors, 56.3% of the nurses recognised that cannulation of a patient aged 60 or above is a risk factor due to older people having more fragile vessels and therefore are more likely to develop phlebitis. This result aligns with the findings of Rego Furtado's (2011b) study, which found that phlebitis was present in approximately 50% of patients over the age of 60. In the present study, 64.3% of the nurses were aware of the increased risk of phlebitis among patients with diabetes mellitus. Studies by both Rego Furtado (2011) and Lima et al (2016) found that patients with diabetes mellitus were more likely to develop phlebitis. Similarly, the nurses' recognition of increased risk of phlebitis from cannulation of patients with ongoing infection was in accordance with Osei-Tutu et al (2015), who found that phlebitis rates were significantly higher among patients with ongoing infections compared to those without such infections, possibly due to the spread of microorganisms by septic emboli from the origin of infection to the cannula tip. This finding is also consistent with Mermel et al's (2009) clinical practice guidelines.

Strengths and limitations

Using simple random sampling was a strength because each nurse in the large population set had the same probability of being selected. The self-administered survey questionnaire, proved to have good reliability with a Cronbach's alpha value of 0.84. A limitation to this study was that it was performed in one tertiary teaching hospital, and thus the results might not reflect the situation across the country. Also, the fact that the majority of participants were female with diplomas in nursing may limit the generalisability of the results. The authors suggest that future research should aim to recruit a larger sample of nurses from several hospitals with similar demographic factors.

Conclusion

Phlebitis is a common complication in the use of peripheral intravenous catheterisation in hospitalised patients. Although generally good, the nurse participants' phlebitis risk factor knowledge was less than satisfactory in some areas. It is recommended that nurses take part in continuing nursing education tailored to improve their knowledge about peripheral intravenous catheter care.

KEY POINTS

CPD reflective questions