Vascular access care in patients with multimorbidity

HIGHLIGHTS

• Multimorbid patients with DIVA required more catheters than those without DIVA.
• The DIVA group required devices of smaller calibre and more insertion attempts.
• Catheter dwell time was similar but adverse events rate was higher in DIVA group.
• Nearly 20% of catheters were inadequate for the therapy prescribed in all patients.
• The results evidenced the need to improve adherence to clinical practice guidelines.

Highly complex patients require an integrated care approach (case management) due the coexistence of multiple health problems, advanced age, greater functional impairment, higher risk of mortality, and greater use of health care resources.^1,2,3 The frequent hospital admissions in this patient population imply multiple intravenous therapies⁴ and higher risk of complications,⁵ including infiltration, phlebitis, thrombosis, and bacteraemia.^6,7,8,9,10 The increased requirement of intravenous therapy and population aging are important contributors to the emergence of difficult venous access (DIVA).^11,12,13

DIVA has been defined as the lack of visible or palpable veins or history of difficulty with catheter placement.^{14,15,16,17,18,19,20} This difficulty leads to an increase in the number of unsuccessful attempts at catheter placement, the use of inappropriate catheter insertion sites,^9,11 delays in administration of medication, and loss of medication through leakage. Further, patients are more likely to need central venous catheters, which are associated with a greater risk of serious complications.¹¹

Numerous studies have investigated the safety of vascular access care through the assessment of the appropriateness of device selection depending on the drug that needs to be administered and anticipated therapy duration, catheter insertion site, number of attempts, catheter type and size, securing of the device, appropriateness of intraluminal care, complication rate, and catheter dwell time.^{4,8,10,21,22,23,24,25} Nevertheless, to the best of our knowledge, no previous studies have focused on patients with multimorbidity.

To establish effective interventions that maximize the safety and success of the intravenous therapy in this population, the prevalence of DIVA should be addressed, and the vascular access care provided should be described. In a previous study, the prevalence of DIVA was determined by assessing the presence of DIVA in patients with multimorbidity at the time of admission, regardless of the type of intravenous therapy or type of catheter expected to be used for treatment purposes.²⁶ Building on the previous study, the objectives of the present study were:

• To describe vascular access care provided to patients with multimorbidity in relation to choice and placement of devices, catheter dwell time, and occurrence of adverse events.
• To explore the differences in the above mentioned factors between the easy and DIVA groups.

Methods

Study design and setting

A prospective cohort study was conducted in 2 tertiary hospitals between March and November 2013. Both hospitals belong to the Public Health Service of our region. The study was carried out across all inpatient wards to which eligible patients were admitted, both medical and surgical (excluding the emergency departments and intensive care units). The follow-up period was from the day of admission to the day the patient was discharged.

Study participants

The target population were all highly complex patients, according to the Kaiser Permanente triangle, who were admitted to hospital during the study period. The patients met the following 3 classification criteria established by the Department of Health:

• Having 3 or more chronic illnesses,
• Having been hospitalized at least once in the previous year, and
• Obtaining a score of 6 to 12 on the Predictive Index of Healthcare Resources Use.

Patients unable to give written informed consent and those previously cared for by an intravenous therapy team (ITT) were excluded.

At recruitment, patients were assigned into the following 2 groups on the basis of an interview with patients and/or family members and physical assessment:

• Easy venous access group: composed of patients who met the following 2 criteria: (1) presence of visible or palpable veins, after tourniquet placement, and (2) no history of more than 2 attempts being required to place an intravenous device.
• DIVA group: composed of patients who did not meet both aforementioned criteria.

Considering an expected prevalence rate of DIVA of 60%, an accuracy of 7%, and an α error of 0.05, the estimated sample size was 133 patients in total. Patients were recruited by consecutive sampling.

Procedure and data collection

Patients were recruited on admission by case management nurses who informed the research team. A member of the research team then visited each patient and sought his or her written informed consent. Once patients had given consent, they were allocated into 1 of the study groups after DIVA assessment and baseline data, demographic, and study variables were collected. The assessment of DIVA and data collection was based on physical examination, interviews with patients, family members or assigned nurses, and review of medical records. From then on, the research team members, in coordination with ward nurses, collected data in a logbook daily until completion of the follow-up period (Table 1).

In the 2 participating hospitals, the venous access devices available—peripheral, midline, and central catheters—were the same in terms of diameter and length. All catheters were placed by ward nurses except for the cases that required additional technology such as ultrasound guidance due to presence of DIVA and those that required midline or central venous catheters. In such cases, catheters were placed by the ITT and intensive care physicians, respectively. All ward nurses followed the same protocol currently in place in the hospitals for catheter placement (blind insertion) and catheter care, which followed the recommendations of clinical practice guidelines.^8,10 The participating ward nurses had basic training on insertion and care of intravenous catheters and on the selection of devices and insertion sites.

The protocol for the placement of short peripheral catheters included disinfecting the skin with chlorhexidine, inserting the device under aseptic conditions, and securing the line with a transparent polyurethane dressing. Intraluminal care involved disinfecting the needleless connector before using, checking catheter patency flushing with saline before and after administration of medication, and changing the needleless connector once a week, while extraluminal care involved disinfecting the skin with chlorhexidine and changing the dressing once a week.

Placement of a midline or central venous catheter was based on maximal sterile barriers, and as for other catheters, the line was secured with a transparent dressing. Intraluminal care involved flushing with saline and a dose of heparin solution and changing the needleless connector once a week, or in the case of catheters used intermittently, flushing after administration of the medication. Extraluminal care was as described above for short peripheral catheters.

Data analysis

The vascular access care provided was described by means of frequencies and percentages for qualitative variables and measures of central tendency and dispersion for quantitative variables. The incidence of adverse events was expressed as number per 1000 catheter days. Secondly, to compare the care provided between the difficult and easy access groups, the χ² test was used for qualitative variables and Student's t test or the Mann-Whitney U test for quantitative variables. Thirdly, to determine the risk of complications associated with DIVA, odds ratios (ORs) were calculated. For all analyses, the level of significance was set at α = 0.05, and 95% confidence intervals were calculated. Lastly, catheter survival was calculated from the day of insertion to the day of removal, which could occur as a result of the end of treatment, death, requirement of central venous catheter, or occurrence of adverse events. The factors explored in the survival analysis were those identified as risk factors for DIVA in a previous study²⁶ and shown in Table 1.

Data were analyzed using IBM SPSS Statistics for Windows, Version 22 (Company, City, ST).

Ethics

All participating patients gave written informed consent. Anonymity and data confidentiality were maintained in accordance with the current local legislation and the Declaration of Helsinki. The study was approved on the 13 February 2013 by the relevant Hospital Clinical Research Ethics Committee.

Results

A total of 135 patients with multimorbidity were recruited. The mean age was 78.38 years (standard deviation [SD] = 9.34), and 34.8% were women. The results of the variables that characterized the vascular access care provided to all patients are summarized in Table 2, and the distribution of patients in both groups is summarized in Table 3.

Statistically significant differences were observed between the groups for the following variables related to vascular access care, namely: the use of small size catheters (22G; P = 0.002), more than 2 cannulation attempts (P < 0.001), and the presence of hematomas at the time of insertion (P = 0.007). Further, although differences were not statistically significant, the DIVA group required more catheters, had shorter dwell times, and experienced more adverse events (see Table 3).

The incidence of adverse events was 58/522 (111.1/1000 catheter days; 95% CI = 84.4 to 143.6) and 28/355 (83.6/1000 catheter days; 95% CI = 55.5 to 120.8) in the difficult and easy access groups, respectively, but such difference was not statistically significant (P = 0.258; see Table 4). There were 59.6% more adverse events in the DIVA group as compared to the easy access group (OR = 1.596; 95% CI = 0.904 to 2.816).

Regarding the type of adverse events (Figure 1), certain events occurred in the DIVA group but were not observed in the easy access group, such as hematoma at the catheter insertion site, pain during the administration of medication, and slow flow rate (Table 4).

Figure 1. Description of the occurrence of adverse events by study group.

It should be noted that 21.95% of the catheters were inappropriate for prescribed therapy, mainly as a consequence of irritating and hyperosmolar substances, 97.7% (43/49) out of the total number of inappropriate devices. Nevertheless, in these devices, the incidence of complications in the DIVA group was 51.7% (15/29) versus 35% (7/20) in the easy venous access group, although this difference was not statistically significant (P = 0.247). Concerning the potential risk factors for DIVA considered in the survival analysis, age was found to be significant. In this regard, the percentage of catheters that functioned well until removal was greater in patients under 75 years of age from day 2 (P = 0.032). All other factors assessed in the survival analysis that have previously been associated with DIVA did not show statistical significance.

Discussion

To the best of our knowledge, no previous studies have assessed the vascular access care provided to patients with multimorbidity, although several studies have been conducted in other patient populations, generally in relation to short peripheral venous catheters,^{27,28,29,30,31,32} which assessed similar aspects to the ones addressed in the present study.

Regarding the catheter type, short peripheral venous catheters were the most commonly used in both groups. According to clinical practice guidelines, short catheters are the most appropriate type of catheter for therapies lasting less than 6 days, assuming that the drugs administered are not irritant or hyperosmolar substances.^8,10,21,25 The use of these types of drugs has been linked to a higher rate of complications.^{27,28,31,32,33,34,35} The survival analysis did not show statistically significant differences between the groups in terms of inappropriateness of the catheter for the prescribed therapy. However, the result of the study showed that irritant substances were used in a substantial percentage of cases, approaching 20% in both study groups. These findings highlight the need to promote adherence to clinical practice guidelines in the selection of the catheter type according to the drugs administered.

Regarding catheter size, the most widely used was 20G, but smaller devices (ie, 22G) were used significantly more frequently in patients with DIVA. Small gauge catheters have been associated with less endothelial damage and, in some studies, with a lower incidence of complications.^{13,28,29,31,33,36,37,38,39,40,41} However, smaller catheters were not always the first choice,³³ as reflected in the present study where the selection of smaller devices was related to the limited venous access in the patients. Hence, the study findings underline the need for assessment of the catheter size and the diameter of the vein previous to the administration of intravenous therapy independently of whether patients have easy venous access or not.

In the group with DIVA, there was a higher rate of insertion of catheters at inappropriate sites (nearly 50%), which was associated with a shorter dwell time in the literature. Notably, the back of the hand was used as the insertion site twice as often in the difficult access group. Previous studies have considered catheter insertion site to be predictive of complications and a key factor in the safety of vascular access care,^{8,27,29–32,38,32,42,43,44} particularly the back of the hand²⁹ and the cubital fossa.^30,33,35 In this regard, there seems to be consensus on the importance of the careful choice of the insertion site to improve vascular access care.⁴⁵ Some authors have correlated a larger number of insertion attempts with DIVA.^{14,19,20,46,47,48,49}

The present study supports the same view, since the group with DIVA required significantly more attempts for successful catheter insertion. Another significant variable, hematoma, was twice as common in patients belonging to the DIVA group (24.8% versus 10.3%). This fact may be indicative of poor management of venous access. We believe that the presence of hematoma should be among the key factors for the ITT when assessing and selecting devices. Catheter dwell time was around 3 days, which was clearly shorter than that expected for these devices and was consistent with data found in the literature.^27,28,31 The dwell time was shorter in the DIVA group (3.84 versus 3.99), although the difference did not reach statistical significance.

Most catheters were inserted by ward nurses, and the ITT only got involved when ultrasound-guided placement was judged necessary. Greater involvement of the ITT might have resulted in a more appropriate choice of the insertion site and catheter type and size, as well as in a reduction of the number of insertion attempts.^47,49

The overall incidence of adverse events was higher than that found in other studies,²⁷ and although differences between groups were not statistically significant, the occurrence of adverse events may be relevant for clinical practice given that patients in the DIVA group were 1.6-fold more likely to have an adverse event. The most common adverse events found in the present study were in line with those reported in the literature.^{8,10,23,24,36} The incidence of phlebitis was lower than that reported in other studies,^{27,30,32,35,50} as well as the incidence of infiltration.³⁰

Secondary outcomes concerning vascular access care may be useful for the identification of patient populations with a greater risk of exhaustion of vascular access sites. First, the results derived from the present investigation indicate a significantly higher risk in people aged 75 years and above,⁴⁹ while there is a lack of consensus in the literature regarding the age-associated risk. Specifically, old age has been associated with more fragile veins^7,19,51 and with a higher rate of complications,⁵² though some studies have failed to demonstrate such associations.^27,32,33 Second, a number of previous studies showed a higher risk in women.^{9,27,29,30,31,52} In the present study, the difference between sexes did not reach statistical significance; however, catheters were more successful in men.

Limitations of the study

A nonprobability sampling and seasonal variability during short recruitment periods might have introduced bias in the study. To account for this limitation, the recruitment period was extended to ensure the inclusion of months with higher and lower rates of admissions of patients with multimorbidity.

Interviewer and memory bias may have been present during group assignment, as data were collected on the item “more than 2 attempts required for the insertion of 1 catheter” referred to the past. To reduce these sources of bias, researchers were appropriately trained, and patients' health records were reviewed.

Given the pragmatic nature of the present study, ward nurses were mainly in charge of the vascular access care of the study patients, and although they all had basic knowledge in vascular care, the actual experience of the nurses inserting catheters was not assessed.

Conclusions

The present study explored the differences in ward nurses' approach to catheter usage and placement in multimorbid patients depending on whether patients presented with DIVA or not. Multimorbid patients with DIVA required more and smaller vascular devices, were more likely to develop hematoma during catheter insertion, and required more catheterization attempts. The study showed that vascular access should be improved in patients with multimorbidity by means of ITTs and by developing appropriate strategies to improve vascular access in this patient population.

Recommendations for practice

The findings may help clinicians identify high-risk populations in terms of vascular access. In order to guarantee a successful intravenous therapy in this patient population, ITTs could be a relevant resource to identify early and accurately DIVA and therefore insert the most suitable catheter for each patient and with the fewest attempts to preserve vascular access.