Peripheral intravenous catheterization is the most frequently used invasive procedure and is usually performed by nurses in contemporary medical practice. Every year, more than a billion peripheral intravenous catheters (PIVCs) are used globally for hospitalized patients. However, PIVC prevalence and management data are lacking from low- and middle-income countries (LMIC) (Alexandrou et al, 2015). Three-fourths of all patients admitted to tertiary care have a vascular device in place mainly for medication therapies, fluid resuscitation, nutritional supplements, or blood and blood products transfusions (Keogh et al, 2015; Mihala et al, 2018).
Peripheral intravenous catheter insertion exposes patients to a number of early or delayed complications, minor or major based on the intensity of symptoms (Salma et al, 2019).
Among the major complications, phlebitis, caused by the inflammation of the intimal layer of the vein, is found to be the most common, leading to the removal of almost half of PIVCs (Braga et al, 2016; Suliman et al, 2020).
While the acceptable rate of phlebitis in any population of patients is at most 5% (Daud, 2018; Gorski et al, 2021), studies conducted in different parts of the world revealed varying findings of PIVC-induced phlebitis: 11.09% in Portugal (Salgueiro-Oliveira et al, 2012), 21.3% in France (Miliani et al, 2017), 31.1% in Brazil (Enes et al, 2016), 59.1% in Nepal (Singh et al, 2008), and 60% in India (Gupta et al, 2007). Peripheral intravenous catheter complications cause prolonged hospitalization and increased cost of treatment (Sari et al, 2016). A study of intensive care units in hospitals in USA reported an increase from 7 to 20 days in hospitalization and additional costs of treatment up to $56,000, due to PIVC-related complications each year (Helm et al, 2015).
Existing literature reported an association of PIVC-induced phlebitis with older age, (Decker et al, 2016), female gender (Rickard et al, 2013), and medical conditions that cause circulatory alterations (Decker et al, 2016). Also, the size of the PIVC is recognized as a risk factor for the development of phlebitis (Mandal and Raghu, 2019). A PIVC with a large diameter raises the risks of phlebitis (Rickard et al, 2013), while a small-gauge PIVC decreases this risk (Milutinovicć et al, 2015). An inappropriately secured PIVC also increases the risk of complications (Comparcini et al, 2017; Gorski et al, 2021). The nature of infusate (eg, irritant, vesicant), frequency of infusion (periodic versus continuous), and catheter site (eg area of flexion as a high-risk location for developing traumatic phlebitis) also influence the occurrence of PIVC-induced phlebitis (Suliman et al, 2020).
Nurses play a key role in the PIVC-related procedures in Pakistan. Generally, nurses with little access to peripheral vascular education, training, and opportunities to practise the skill are involved in insertion and caring of PIVCs in routine daily patient care (Schuster et al, 2016). We believe that lack of robust, standardized training processes for this invasive procedure led to clinical performance variability and resultant procedural failures, including increased catheterization attempts, clients’ discomfort, and recurrent PIVC failure. Recognition and minimization of adverse events related to intravascular devices (Russell et al, 2014) and PIVC-related complications are treated as a nurse-sensitive quality indicator in healthcare settings (Salgueiro-Oliveira et al, 2019).
Studies conducted in different parts of the world revealed varying findings of the incidence of PIVC-induced phlebitis and their associated factors. Nonetheless, data from LMIC is scarce, especially from Pakistan. A few studies conducted in Pakistan have investigated the factors contributing to the development of phlebitis (Bherulal et al, 2020; Bibi et al, 2023; Khoso et al, 2021; Shahnaz et al, 2021), phlebitis among postoperative patients (Shahnaz et al, 2021), and knowledge regarding risk factors of phlebitis and its association with educational level among nurses (Bibi et al, 2023). However, we have not found such a study that prospectively investigates the development of PIVC-induced phlebitis and relates PIVC-induced phlebitis with the level of care of PIVCs in Pakistan. Therefore, we aimed to prospectively determine the proportion of PIVC-induced phlebitis and its predictors with an emphasis on care of PIVCs by nurses among the hospitalized patients in a public sector tertiary care hospital in Karachi, Pakistan.
Methods
Study design, setting and participants
We conducted a hospital-based cohort study during March 2019 to May 2019. The study was conducted in the general medical, surgical, gastroenterology, and gynecology wards of the Dow University Hospital, Karachi. Dow University Hospital is a 1000bed tertiary care not-for-profit public sector hospital located in Gulzar-e-Hijri town, in the district east of Karachi, Pakistan. The hospital caters to the population of Karachi and other nearby cities of the country, including patients from both low- and middle-income backgrounds.
Patients aged 18 years and above, admitted in the selected wards, and planned for PIVC insertion were eligible for enrollment in the study. Patients admitted for daycare procedures or hemodynamic monitoring and planned PIVC removal within 24 hours, patients with pre-existing skin breakdowns/lesions at insertion sites (rashes, lacerations, burns, and trauma), immunocompromised patients (such as patients with HIV/AIDS, cancer, and liver/renal transplant), and those receiving chemotherapy were excluded. The principal investigator (PI) identified the potentially eligible participants on a daily basis from the daily census sheet located in each unit and approached the patients for further screening and seeking written informed consent for participation into the study.
Sample size calculation was performed using OpenEpi online software. Considering the incidence of PIVC-induced phlebitis as 21.3% (Miliani et al, 2017) among the hospitalized patients, assuming 5% bound on error of estimation, and a 95% confidence interval, the required minimum sample size for this study was 258 (n=258).
Data collection tools and procedures
Data was collected using two tools. The first tool included two self-developed questionnaires comprising a) patient-related characteristics (Appendix A*) and b) PIVC-related characteristics (Appendix B*). This tool was developed based on an extensive review of the previously published literature (Abolfotouh et al, 2014; Keogh et al, 2015; Salgueiro-Oliveira et al, 2012; Souza et al, 2015) and was reviewed and approved by a group of subject specialists, including nurses and medical practitioners. The second tool, Structured Observation Protocol (Appendix C*), was chosen because it includes components to assess simultaneously 1) the phlebitis scale and 2) care of PIVCs. This allowed better monitoring for PIVC-induced phlebitis (Ahlqvist et al, 2006). This tool was validated by Ahlqvist et al (2006) in Sweden and we used it, as it was, for data collection in this study.
Using the self-developed questionnaires (Appendices A and B*), data concerning patients and PIVC intravenous therapy were collected by the PI through observations of PIVC, medical records review, and questioning the patients. Through the Structured Observation Protocol, the PI met the admitted patients who had a PIVC in place for direct observation of PIVC-induced phlebitis and PIVC care.
Grading of these observations was carried out based on the protocol. Patients were asked for certain responses, such as the number of attempts during the current PIVC insertion and any pain at the PIVC insertion site; other observations at the PIVC site were carried out by the PI. Each patient who had a PIVC inserted within the last 24 hours, was assessed daily (once in each 24-hour period) in the same unit, until removal of the PIVC for selected items of the study instrument, such as PIVC-induced phlebitis, care of the PIVC (Appendix C*), number and type of intravenous medicines and fluids, and blood transfusion.
Ethical consideration
Permission was obtained from the hospital administration and ethical approval was obtained from the institutional review board of Dow University of Health Sciences, Karachi, Pakistan (Ref: IRB-1190/DUHS/Approval/2019/04).
Written informed consent was obtained prior to the recruitment of the participants. All data were collected using anonymous questionnaires. All paper form data were kept within a locked cabinet. Electronic data were kept in a password-protected secured computer at the PI's office. Data were not shared with anyone outside the study team.
Data analysis
Data entry and statistical analysis were performed using the STATA version 16.0. Frequency with percentage was calculated for all categorical variables reported in Table 1. Comparison of sociodemographic, clinical, and PIVC-related characteristics between participants with phlebitis and without phlebitis were performed using Pearson chi-squared or Fisher exact tests. The P value (P≤0.05) was set as the criterion for statistical significance. Univariate and multivariate logistic regression was performed to identify the predictors of phlebitis. Variables with P values of less than 0.20 at the univariate level were moved to multivariate logistic regression analysis. Best subset method was used for the selection of variables in the multivariate logistic regression model. Two-sided P values of 0.05 were considered significant in the multivariate logistic regression model. Adjusted odds ratios with 95% confidence intervals (95% CI) were calculated.
| Characteristics | Phlebitis | P value | |||
|---|---|---|---|---|---|
| Total, n=258 | Yes, n=101 (39.1%) | No, n=157 (60.9%) | |||
| Age group (years) | 20–40 | 104 (40.3) | 42 (40.4%) | 62 (59.6%) | 0.485 |
| 41–60 | 77 (29.8) | 26 (33.80%) | 51 (66.20%) | ||
| 61–80 | 77 (29.8) | 33 (42.90%) | 44 (57.10%) | ||
| Gender | Male | 119 (46.1) | 44 (37.0%) | 75 (63.0%) | 0.297 |
| Female | 139 (53.9) | 57 (41.0%) | 82 (59.0%) | ||
| Comorbidity | |||||
| 1. DM | No | 175 (67.8) | 70 (40.0%) | 105 (60.0%) | 0.395 |
| Yes | 83 (32.2) | 31 (37.3%) | 52 (62.7%) | ||
| 2. HTN | No | 165 (63.9) | 68 (41.2%) | 97 (58.8%) | 0.220 |
| Yes | 93 (36.1) | 33 (35.5%) | 60 (64.5%) | ||
| 3. IHD | No | 243 (94.2) | 93 (38.3%) | 150 (61.7%) | 0.187 |
| Yes | 15 (5.8) | 8 (53.3%) | 7 (46.7%) | ||
| 4. Asthma | No | 257 (99.6) | 100 (38.9%) | 157 (61.1%) | 0.391 |
| Yes | 1 (0.4) | 1 (100%) | 0 | ||
| 5. TB | No | 249 (96.5) | 93 (37.3%) | 156 (62.7%) | 0.003* |
| Yes | 9 (3.5) | 8 (88.9%) | 1 (11.1%) | ||
| 6. Others | No | 243 (94.2) | 98 (40.3%) | 145 (59.7%) | 0.095 |
| Yes | 15 (5.8) | 3 (20.0%) | 12 (80.0%) | ||
| PIVC insertion shift | Morning shift | 118 (45.7) | 46 (39.0) | 72 (61.0%) | 0.653 |
| Evening shift | 62 (24.0) | 27 (43.5%) | 35 (56.5%) | ||
| Night shift | 78 (30.2) | 28 (35.9%) | 50 (64.1%) | ||
| Catheterized limb | Right upper | 138 (53.5) | 55 (39.9%) | 83 (60.1%) | 0.452 |
| Left upper | 120 (46.5) | 46 (38.3%) | 74 (61.7%) | ||
| PIVC insertion site | Hand | 80 (31.0) | 24 (30.0%) | 56 (70.0%) | 0.020* |
| Wrist | 65 (25.2) | 35 (53.8%) | 30 (46.2%) | ||
| Forearm | 94 (36.4) | 33 (35.1%) | 61 (64.9%) | ||
| Antecubital area | 19 (7.4) | 9 (47.4%) | 10 (52.6%) | ||
| PIVC size (gauge) | 18 | 46 (17.8) | 15 (32.6%) | 31 (67.4%) | 0.610 |
| 20 | 148 (57.4) | 60 (40.5%) | 88 (59.5%) | ||
| 22 | 64 (24.8) | 26 (40.6%) | 38 (59.4%) | ||
| PIVC stabilization material | Adhesive tape | 147 (57.0) | 61 (41.5%) | 86 (58.5%) | 0.224 |
| Transparent dressing | 111 (43.0) | 40 (36.0%) | 71 (64.0%) | ||
| Number of attempts | 1 | 211 (81.8) | 78 (37.0%) | 133 (63.0%) | 0.540 |
| 2 | 44 (17.1) | 20 (45.5%) | 24 (54.5%) | ||
| 3 | 3 (1.2) | 3 (100%) | 0 | ||
| Number of IV medicines | 0–02 | 59 (22.9) | 17 (28.8%) | 42 (71.2%) | 0.177 |
| 03–04 | 124 (48.1) | 53 (42.7%) | 71 (57.3%) | ||
| 05–06 | 75 (29.1) | 31 (41.3%) | 44 (58.7%) | ||
| Types of IV medicines | |||||
| 1. Antibiotics | No | 24 (9.3) | 10 (41.7%) | 14 (58.3%) | 0.476 |
| Yes | 234 (90.7) | 91 (38.9%) | 143 (61.1%) | ||
| 2. GI drugs | No | 47 (18.2) | 13 (27.7%) | 34 (72.3%) | 0.051* |
| Yes | 211 (81.8) | 88 (41.7%) | 123 (58.3%) | ||
| 3. CV drugs | No | 244 (94.6) | 96 (39.3%) | 148 (60.7%) | 0.512 |
| Yes | 14 (5.4) | 5 (35.7%) | 9 (64.3%) | ||
| 4. Analgesics | No | 153 (59.3) | 59 (38.6%) | 94 (61.4%) | 0.458 |
| Yes | 105 (40.7) | 42 (40.0%) | 63 (60.0%) | ||
| 5. Antipyretics | No | 180 (69.8) | 61 (33.9%) | 119 (66.1%) | 0.007* |
| Yes | 78 (30.2) | 40 (51.3%) | 38 (48.7%) | ||
| 6. Others | No | 208 (80.6) | 81 (38.9%) | 127 (61.1%) | 0.507 |
| Yes | 50 (19.4) | 20 (40.0%) | 30 (60.0%) | ||
| IV fluid therapy | No | 74 (28.7) | 20 (27.0%) | 54 (73.0%) | 0.008* |
| Yes | 184 (71.3) | 81 (44.0%) | 103 (56.0%) | ||
| Frequency of infusion | Continuous | 182 (70.5) | 81 (44.5%) | 101 (55.0%) | 0.312 |
| Intermittent | 2 (0.8) | 0 | 2 (100%) | ||
| Type of IVF | |||||
| 1. 0.9% N/S | No | 63 (24.4) | 27 (42.9%) | 36 (57.1%) | 0.472 |
| Yes | 121 (46.9) | 54 (44.6%) | 67 (55.4%) | ||
| 2. 0.45% N/S | No | 172 (66.7) | 75 (43.6%) | 97 (56.4%) | 0.444 |
| Yes | 12 (4.7) | 6 (50.0%) | 6 (50.0%) | ||
| 3. 5% D/W | No | 169 (65.5) | 73 (43.2%) | 96 (56.8%) | 0.311 |
| Yes | 15 (5.8) | 8 (53.3%) | 7 (46.7%) | ||
| 4. 5% D/S | No | 179 (69.4) | 77 (43.0%) | 102 (57.0%) | 0.118 |
| Yes | 5 (1.9) | 4 (80.0%) | 1 (20.0%) | ||
| 5. Lactated ringers | No | 137 (53.1) | 60 (43.8%) | 77 (56.2%) | 0.524 |
| Yes | 47 (18.2) | 21 (44.7%) | 26 (55.3%) | ||
| 6. Amino acids | No | 180 (69.8) | 79 (43.9%) | 101 (56.1%) | 0.594 |
| Yes | 4 (1.6) | 2 (50.0%) | 2 (50.0%) | ||
| Additional supplements in the IVF | Nil | 160 (62.0) | 67 (41.9%) | 93 (58.1%) | 0.299 |
| Potassium chloride | 11 (4.3) | 6 (54.5%) | 5 (45.5%) | ||
| Others | 13 (5.0) | 8 (61.5%) | 5 (38.5%) | ||
| Blood/blood product transfusion | No | 249 (96.5) | 97 (39.0%) | 152 (61.0%) | 0.497 |
| Yes | 9 (3.5) | 4 (44.4%) | 5 (55.6%) | ||
| Total PIVC dwell time (hours) | Less than 24 | 3 (1.2) | 3 (100%) | 0 | <0.001* |
| 24.1–48.0 | 22 (8.5) | 21 (95.5%) | 1 (4.5%) | ||
| 48.1–72.0 | 165 (63.9) | 49 (29.7%) | 116 (70.3%) | ||
| 72.1–96.0 | 46 (17.8) | 20 (43.5%) | 26 (56.5%) | ||
| More than 96.0 | 22 (8.5) | 8 (36.4%) | 14 (63.6%) | ||
Note *P value <0.05(Chi-square test) was considered significant;
DM=diabetes mellitus, HTN hypertension, IHD=ischemic heart disease, TB=tuberculosis, PIVC=peripheral intravenous catheter, IV=intravenous, GI=gastrointestinal, CV=cardiovascular, VF=intravenous fluid, N/S=normal saline, D/W=dextrose in water, D/S=dextrose in sodium chloride
Results
A total of 258 patients who had a PIVC in place and were admitted in various units were enrolled in the study and followed until the removal of their PIVC. The mean age of the patients was 47.2±17.7 years. Many patients (n=104; 40.3%) were young adults in the age group of 20–40 years and of the 258, the majority (n=139; 53.9%) were females. The majority (165; 64.0%) of the PIVCs indwelled between 48.1 and 72.0 hours, while 22 (8.5%) remained in place for more than 96.0 hours. Most of the PIVCs (n=157; 60.9%) were removed due to therapy completion or discharge of patients, while 101 (39.1%) were removed due to development of phlebitis. A total of 101 (39.1%) patients developed phlebitis during the hospitalization. Comparison of the sociodemographic, clinical, and PIVC-related characteristics between those who developed phlebitis and those who did not develop phlebitis were statistically similar except for tuberculosis (TB) as a comorbidity (P=0.003), PIVC insertion site (P=0.02), intravenous (IV) administration of gastrointestinal drugs (P=0.051), IV antipyretics (P=0.007), and PIVC dwell time (P< 0.001; Table 1). Complete details of sociodemographic, clinical, and PIVC-related characteristics of the participants are presented in Table 1.
Day-wise development of phlebitis is illustrated in Table 2, which reveals that on Day 0, out of 258, only 3 (1.2%) cases were found with light/medium phlebitis. On Day 1, out of the remaining 255 patients, 21 (8.2%) were reported with light/medium phlebitis and 2 (0.8%) with severe phlebitis. Among the other 234 patients, most of the cases (53; 22.6%) of phlebitis were found on Day 2 in various degrees: light/medium (45; 19.2%), severe (6; 2.6%), and very severe (2; 0.9%). On Day 3, out of 69 patients, 19 developed phlebitis: light/medium (17; 24.6%), severe (1; 1.4%), and very severe (1; 1.4%). Day 4 findings revealed that 8 (33.3%) of the remaining 24 patients, had light/medium phlebitis. On Day 5, none of the six remaining patients developed phlebitis. Then, on Day 6, of the remaining 3 patients, 1 (33.3%) was found with light/medium phlebitis.
| Day | Phlebitis status | Frequency (n) | Percentage |
|---|---|---|---|
| Day 0 | No sign | 255 | 98.8% |
| Light/medium | 3 | 1.2% | |
| Severe | 0 | 0% | |
| Very severe | 0 | 0% | |
| Total | 258 | 100% | |
| Day 1 | No sign | 232 | 91% |
| Light/medium | 21 | 8.2% | |
| Severe | 2 | 0.8% | |
| Very severe | 0 | 0% | |
| Total | 255 | 100% | |
| Day 2 | No sign | 181 | 77.4% |
| Light/medium | 45 | 19.2% | |
| Severe | 6 | 2.6% | |
| Very severe | 2 | 0.9% | |
| Total | 234 | 100% | |
| Day 3 | No sign | 50 | 72.5% |
| Light/medium | 17 | 24.6% | |
| Severe | 1 | 1.4% | |
| Very severe | 1 | 1.4% | |
| Total | 69 | 100% | |
| Day 4 | No sign | 16 | 66.7% |
| Light/medium | 8 | 33.3% | |
| Severe | 0 | 0% | |
| Very severe | 0 | 0% | |
| Total | 24 | 100% | |
| Day 5 | No sign | 6 | 100% |
| Light/medium | 0 | 0% | |
| Severe | 0 | 0% | |
| Very severe | 0 | 0% | |
| Total | 6 | 100% | |
| Day 6 | No sign | 2 | 66.7% |
| Light/medium | 1 | 33.3% | |
| Severe | 0 | 0% | |
| Very severe | 0 | 0% | |
| Total | 3 | 100% |
Multivariate logistic regression demonstrated that TB as a comorbidity, PIVC dwell time before initial assessment, administration of IV fluids, and unsatisfactory PIVC care at Day 1 were significant predictors for the development of phlebitis among the enrolled participants. The adjusted odds of developing phlebitis were 7.6 times more likely among those with TB, as compared to those without TB as a comorbidity (Adj. OR 7.59; 95% CI 1.31–44.20). Likewise, there was a dose response relationship between the PIVC dwell time before initial assessment (in hours) and the development of phlebitis. In comparison to a less-than-six-hour PIVC dwell time before initial assessment, the adjusted odds ratio of developing phlebitis was 1.95; 95% CI 0.58–6.60 for those with a PIVC dwell time of 6 to 12 hours, 4.49; 95% CI 1.48–13.58 for those with a PIVC dwell time of 13 to 18 hours, and 5.39; 95% CI 1.98–14.72 for those with a PIVC dwell time of 19 to 24 hours. Similarly, the adjusted odds of developing phlebitis were 3.04; 95% CI 1.50–6.18 among those receiving IV fluids as compared with those not receiving IV fluids and, among those who received dissatisfactory PIVC care after one day of hospitalization, the adjusted odds of phlebitis were 8.65; 95% CI 4.37–17.11 as compared with those who received satisfactory PIVC care (Table 3).
| Variables | cOR (95% CI) | Adj. OR (95% CI) | P values for adj. OR | |
|---|---|---|---|---|
| TB as comorbidity | 5.77 (1.17–28.36) | 7.59 (1.31–44.20) | 0.024 | |
| Other comorbidities* | 0.37 (0.10–1.34) | – | ||
| PIVC dwell time (in hours) before | < 6 | Ref | Ref | |
| initial assessment | 6 to 12 | 2.0 (0.75–5.37) | 1.95 (0.58–6.60) | < 0.001 |
| 13 to 18 | 3.13 (1.27–7.73) | 4.49 (1.48–13.58) | ||
| 19 to 24 | 2.46 (1.12–5.40) | 5.39 (1.98–14.72) | ||
| PIVC insertion site | Hand | Ref | ||
| Wrist | 2.72 (1.38–5.39) | |||
| Forearm | 1.26 (0.67–2.39) | |||
| Antecubital | 2.10 (0.76–5.82) | |||
| PIVC stabilization material | Adhesive tape | Ref | ||
| Transparent dressing | 0.91 (0.55–1.51) | |||
| Total number of IV medications | 1.16 (0.97–1.37) | |||
| IV antibiotics | 1.08 (0.45–2.57) | |||
| IV cardiovascular drugs | 0.86 (0.28–2.63) | |||
| IV analgesics | 1.06 (0.64–1.77) | |||
| IV GI drugs | 1.65 (0.84–3.27) | |||
| IV antipyretics | 2.05 (1.19–3.53) | |||
| Receiving IV fluids | 2.33 (1.28–4.23) | 3.04 (1.50–6.18) | 0.002 | |
| Unsatisfactory care at day 0 | 10.39 (2.95–36.47) | – | – | |
| Unsatisfactory care at day 1 | 6.67 (3.65–12.13) | 0.002 | <0.001 | |
| Unsatisfactory care at day 2 | 37.42 (17.27–81.06) | – | – | |
| Unsatisfactory care at day 3 | 37.42 (17.27–81.06) | – | – |
Note. *Other than diabetes mellitus, ischemic heart disease, asthma, hypertension, tuberculosis, odds ratios are calculated using binary logistic regression (univariate followed by multivariate logistic regression). Adjusted odds ratios are adjusted for the variables in the model and PIVC insertion site, and PIVC size (irrespective of their statistical non-significance)
GI=gastrointestinal, IV=intravenous, PIVC=peripheral intravenous catheter, TB=tuberculosis
Discussion
Research on PIVC-induced phlebitis and its associated factors in Pakistan and regional countries is limited. This study, with a robust research design, adequate sample size, conducted in a tertiary care public sector hospital, and using a validated tool for the assessment of phlebitis, provides useful data for nursing administrators to develop an evidence-based nursing care plan to prevent PIVC-induced phlebitis among the hospitalized patients in Pakistan.
The proportion of phlebitis in this study was 39.1%, with light/medium phlebitis as the most common, followed by severe phlebitis and very severe phlebitis. The burden of phlebitis in this study was higher than in developed countries, such as 11.09% in Portugal (Salgueiro-Oliveira et al, 2012), and lower than in developing countries, such as 60% in India (Gupta et al, 2007). According to the Infusion Nurses Society (INS), the acceptable benchmark of phlebitis rates among hospitalized patients is 5% (Gorski et al, 2021; Makafi and Marfega, 2017), which is substantially lower than the burden of phlebitis we observed in this study from Pakistan. This study also observed varying degrees of PIVC care. While satisfactory level of care for PIVC was reported in the initial days following PIVC insertion, the trend in level of care was found to be a gradual decline. Patients with a dissatisfactory level of PIVC care at Day 2 were more likely to develop phlebitis as compared with those receiving satisfactory level of care on Day 2 (see Appendix C*: Care Grading Scale). An earlier study from Pakistan, reported sufficient knowledge regarding PIVC care among nurses, yet the PIVC care practices were not consistent with the standard protocols (Qamar et al, 2017). A study from Brazil also reported association between poor PIVC-related nursing care and development of phlebitis, emphasizing the importance of PIVC care for prevention of phlebitis (Souza et al, 2015). Since the healthcare system in Pakistan is already burdened with fewer than the required number of beds in public sector hospitals for the population of Pakistan (Hashmi et al, 2020), a significantly higher proportion of phlebitis prolongs hospitalization and increases cost of treatment. Continuous nursing education for nursing staff, development of care protocols for PIVC to prevent complications, and proper documentation of care are some of the cost-effective strategies for decreasing the existing burden of phlebitis.
While some of the previous studies reported a preponderance of phlebitis among female (Lv and Zhang, 2020) or male patients (Singh et al, 2008), in this study we observed no association between gender and phlebitis. Our finding is consistent with some of the earlier studies conducted in different parts of the world, reporting no association between gender and phlebitis (Ben Abdelaziz et al, 2017; Webster et al, 2015). We also did not observe any association between age and phlebitis. However, a study conducted in Australia reported aged population (age 60 years and above) at a higher risk of developing phlebitis (Decker et al, 2016). Age-related thinning of skin (senile fragile), loss of protective fat layers, depletion of sensations especially touch, pressure, vibration, heat and cold making the skin of elderly patients vulnerable to tear and damage, could be potential reasons for phlebitis among an elderly population.
In this study, patients withTB as a comorbidity were eight times more likely to develop phlebitis as compared to those without TB. While none of the earlier studies reported an association between TB and phlebitis, the authors believe the malnutrition, dehydration, and low immunity caused by TB could increase the probability of phlebitis developing among hospitalized TB patients in Pakistan (Cai et al, 2020). Nurses should pay special attention while caring for patients with TB as a comorbidity and who have a PIVC inserted. Developing a nursing care plan with more frequent PIVC care and close observation for phlebitis in every shift is recommended.
In this study, all patients have PIVCs inserted in the upper limbs. Suliman et al (2020) suggested veins of the upper limbs should be considered, instead of those of the lower limbs, due to the risk of embolisms and phlebitis (Suliman et al, 2020). Statistically significant differences were observed between the specific anatomical site used (hand, forearm, wrist, and antecubital area) and phlebitis. Most of the cases of phlebitis were reported when the PIVC was placed in the wrist, which can be attributed to traumatic phlebitis associated with flexion or high mobility areas. These findings are consistent with the findings of a study conducted in Portugal (Salgueiro-Oliveira et al, 2012).
In this study, the administration of IV fluids (including gastrointestinal and antipyretic drugs) was significantly associated with the development of phlebitis. Contrary to a previous study (Braga et al 2018), where IV administration of antibiotics was reported to increase the likelihood of phlebitis, we found no such association. The nature of antibiotics, ie, vesicant versus nonvesicant, could be the possible explanation for the differences in results found in this study versus Braga et al (2018).
This study observed a dose-response relationship between the time duration from PIVC insertion to the initial assessment and risk of developing phlebitis (see Table 3). The shorter the duration in hours between PIVC insertion and initial assessment, the lower the odds of developing phlebitis. As compared with less than six hours duration, the risk of developing phlebitis was more than five times higher when the duration between PIVC insertion and initial assessment was 19 to 24 hours. An earlier study reported the association between PIVC dwell time and risk of developing phlebitis (Ansel et al, 2017). This finding provides important evidence for nursing practice regarding the care of PIVCs in Pakistan. The initial assessment of PIVCs after insertion must be done within six hours of its insertion to avoid the risk of phlebitis.
Limitations
While this is a well designed, sufficiently powered study conducted among adult patients in a large tertiary care public sector hospital, and the development of phlebitis was directly observed prospectively and recorded using a validated tool, there are a few limitations which need to be considered while interpreting the study's results. First, this was a single-centre study recruiting patients from only one hospital; hence, the findings might not be generalizable to other centres, especially tertiary care private hospitals with different catchment populations and better care protocols. Second, patients were recruited only from the general wards; hence findings cannot be generalized to the patients admitted in critical or intensive care units of the Dow University Hospital or other tertiary care public sector hospitals in Karachi, Pakistan. Patients admitted in intensive care units might have a higher incidence of phlebitis. Third, nurses were not observed during PIVC insertion, which may lead to underreporting of PIVC-induced phlebitis caused by variability in the insertion approach.
Conclusion
The proportion of PIVC-induced phlebitis among adult hospitalized patients in Pakistan is higher (39.1%) than the 5% rate acceptable by the Infusion Nurses Society (Gorski et al, 2021). Delay in the initial assessment after PIVC insertion, along with suboptimum PIVC care and comorbid TB, significantly increase the risk of developing phlebitis. Continuous nursing education, developing standard care plans for PIVCs, and proper documentation of care are recommended.