Surgery is a scary matter for most people, combining fear of the unknown with apprehension about such anticipated procedures as insertion of an IV line. All patients admitted for same-day surgery require IV access. Any venipuncture, including peripheral IV catheterization with a medium-to-large-gauge catheter, can cause some degree of pain. Using local anesthesia prior to IV catheterization has been shown to reduce pain1-3 and anxiety.4 Choices for local anesthesia prior to IV catheterization include a topical lidocaine-prilocaine mixture, intradermal lidocaine, or intradermal bacteriostatic normal saline.5-7
For decades, clinical researchers have recommended using intradermal lidocaine to reduce the discomfort of IV initiation, often calling it the gold standard of local anesthetics.3,8,9 Although intradermal administration may cause the patient slight discomfort during the initial needle insertion, it's expected that the introduction of the local anesthetic will mediate the pain of IV catheterization.7 Unbuffered lidocaine is acidic and often produces a stinging sensation upon injection. While buffering the lidocaine with sodium bicarbonate reduces this discomfort,10 buffered lidocaine isn't commercially available; it's classified as a compounded medication and must be prepared in the pharmacy. In addition, it has a short shelf life. When this study was conducted in 2008, the Food and Drug Administration required buffered lidocaine solution to be replaced weekly. The standard has since been relaxed slightly, allowing the drug to remain on the shelf for up to nine days.11,12
Another option for local anesthesia is intradermal bacteriostatic normal saline, which is preserved with benzyl alcohol, an opium alkaloid that has both antiseptic and anesthetic properties. The evidence supporting the anesthetic effect of benzyl alcohol dates back to 1918, and over the past decade or so some clinicians have proposed that intradermal bacteriostatic normal saline may be used as an alternative to intradermal lidocaine for anesthetic pretreatment prior to venipuncture.13
Of the available local anesthetics, the topical lidocaine-prilocaine mixture is least appropriate for same-day surgery. In addition to being the most costly, it must be applied at least 30 minutes before starting an IV line for best effect, which is seldom realistic in the fast-paced context of same-day surgery.
BACKGROUND
In 2005, as part of a larger pain management initiative, nurses on the same-day surgery unit of a Midwestern community hospital initiated a policy of administering intradermal buffered lidocaine to all nonallergic patients before starting IV lines. (The previous practice had been to start IV lines with no anesthetic pretreatment.) When, in 2007, as a cost-saving measure, the hospital administration called for a changeover from intradermal buffered lidocaine to intradermal bacteriostatic normal saline, the nurses were concerned that this alternative might not be as effective an anesthetic and asked the unit journal club to explore the available evidence on local anesthesia for venipuncture in adults.
Literature review. While much has been published on this topic over the past two decades, no systematic reviews were found that compared the efficacy of the two intradermal solutions in question. Previous studies either failed to use buffered lidocaine as a comparison intradermal solution or had been conducted on samples too small to show significant differences.
Wilson and Martin investigated the viability of using either intradermal 0.9% benzyl alcohol or intradermal 1% buffered lidocaine as an anesthetic pretreatment for procedures, such as laceration repair, that require a more prolonged cutaneous anesthesia than venipuncture does.14 To lengthen the anesthetic effect of the two solutions, they added the vasoconstrictor epinephrine (1:100,000) to each. Their findings indicated that the 0.9% benzyl alcohol and epinephrine mixture was significantly less painful to adults on injection than either the 1% buffered lidocaine and epinephrine mixture or normal saline (not bacteriostatic). The efficacy of either anesthesia for venipuncture, however, wasn't tested.
Overall, published reports indicated that IV catheterization without anesthetic pretreatment is more painful than an intradermal injection of unbuffered lidocaine and that the pain of IV catheterization is reduced after injection of unbuffered lidocaine.2,7,15-17 Janet Brown found that 96% of patients given intradermal unbuffered lidocaine prior to an IV catheterization said they'd want it used as an anesthetic pretreatment for future IV catheterization.2 The sample included both men and women, and mean pain ratings for IV line insertion weren't significantly different between the sexes. Hussey and colleagues compared pretreatment with intradermal unbuffered lidocaine or intradermal normal saline with standard venipuncture technique (no anesthetic pretreatment) and reported that unbuffered lidocaine provided significantly better pain control than the alternatives.16 Windle and colleagues examined the differences between intradermal unbuffered lidocaine, intradermal bacteriostatic normal saline, and no anesthetic pretreatment in a large sample (N = 221) and found IV catheterization was less painful when preceded by either intradermal anesthetic; moreover, there was no significant difference in reported pain between the two anesthetics, although bacteriostatic normal saline was associated with less pain on injection than unbuffered lidocaine.7
Hattula and colleagues compared no anesthetic pretreatment for IV catheterization (the control group) with pretreatment with either intradermal buffered lidocaine or intradermal bacteriostatic normal saline.3 Subjects receiving intradermal buffered lidocaine reported less pain than those receiving bacteriostatic normal saline, and no significant difference in pain was reported between the bacteriostatic normal saline group and the control group, suggesting that intradermal buffered lidocaine is the more effective anesthetic. The study, however, included only 33 subjects (11 receiving intradermal buffered lidocaine, 13 receiving intradermal bacteriostatic normal saline, and nine receiving no anesthetic pretreatment), leaving room for continued speculation. In contrast, Dava Brown concluded that intradermal bacteriostatic normal saline was as effective as intradermal buffered lidocaine in reducing the pain of IV catheter placement in adult patients, although this study had only 47 subjects, and Brown recommended that it be replicated to verify results.1
Having found only two small studies that examined the pain-mediating effects of intradermal bacteriostatic normal saline and intradermal buffered lidocaine,1,3 the unit journal club concluded that existing research provided insufficient evidence to support a practice change and proposed conducting a study of its own.
METHODS
Design. After receiving institutional review board approval, we invited male and female patients between the ages of 18 and 80 from our hospital's same-day surgery unit to participate in the study. To be included, participants had to be able to understand, speak, and read English; have a physician's order for a peripheral IV line; and have the ability to rate and express their pain level. Because earlier studies had established a clear benefit to using local anesthesia prior to IV catheterization, we believed that comparing the test solutions with no anesthetic was unnecessary. Instead, we used a randomized, double-blind, quasiexperimental, two-treatment, parallel design to compare the efficacy of intradermal bacteriostatic normal saline with that of intradermal buffered lidocaine in providing local anesthesia to adult patients prior to IV catheterization.
The difference in mean pain scores was assessed using the t test. An analysis of covariance was performed to test for outcome differences related to age, sex, and race or ethnicity.
It was important that this study be sufficiently powered to provide clear evidence in support of one of the two solutions in question. Assuming an [alpha] of 0.05, a power of 0.80, and a moderate effect size, power analysis suggested a sample size of at least 128. To account for the possibility of participant withdrawal, we continued recruiting patients until 75 were enrolled in each treatment arm. All subjects provided written informed consent prior to participation.
Sample. The initial sample included 150 adults admitted for same-day surgery between February and May of 2008. Two of the 150 subjects were excluded from the study because they required more than one attempt at venipuncture, and multiple needle sticks are known to increase anxiety and may cause pain to be reported inaccurately. Patients with a history of lidocaine allergy and those who'd taken pain medication within the previous four hours were also excluded from the study.
Instrument. The intensity of pain perception was measured with a modified verbal numeric rating scale (VNRS) in which 0 represents no pain and 10 represents the worst pain imaginable. Although pain is a subjective phenomenon, a visual analog scale (VAS) or a VNRS is generally accepted as a valid and reliable method of pain sensitivity measurement in an adult population.18,19 Holdgate and colleagues demonstrated that, although the VAS is considered the gold standard of pain measurement, the VNRS is a reliable alternative in certain contexts and has the advantage of being less complex than the VAS and easier for patients to understand.20 (Holdgate points out, however, that the two aren't interchangeable.)
Both the VAS and the VNRS provide nominal (yes or no), ordinal (rank), and interval (equal distance between intervals) data. In addition, the VAS provides ratio data and an absolute zero point, allowing for more sophisticated statistical analysis. Whether the VNRS provides ratio data depends on the question being asked. Within the context of mild-to-moderate pain, such as that experienced with IV catheterization, VAS and VNRS scores have been shown to trend in the same direction. Under such circumstances, therefore, the VNRS can be said to provide ratio data and is thus suitable for clinical research.21,22
Procedure. All subjects and same-day surgery staff were blinded to the study solution used. On each of 121/2 noncontinuous weeks during the four-month study period, a pharmacist prepared the buffered lidocaine by adding 5 mL of 8.4% sodium bicarbonate to a 50 mL vial of 1% lidocaine and filled each of six 30-gauge tuberculin syringes with 0.2 mL of this solution. Another six syringes were each filled with 0.2 mL of commercially prepared bacteriostatic normal saline. According to a preestablished protocol, the pharmacist affixed numbered labels to the syringes and, on a second set of labels, placed in a documentation file, noted the type of solution corresponding to each number. Twelve packets (six of each solution), each containing a filled syringe and a data collection sheet labeled with the same number, were assembled in the pharmacy and delivered to the same-day surgery unit. The data collection sheet included a VNRS and space to record the participant's age, sex, race or ethnicity, and comments about the procedure, as well as the IV location and size of the catheter inserted.
All patients admitted to the same-day surgery unit during the study period were given information about the study and asked by nursing staff if they were interested in participating. A single same-day surgery RN, skilled in IV catheterization, was trained in the study protocol and performed all protocol functions, further explaining the study to participants, obtaining informed consent, administering the study solution, performing the venipuncture, administering the pain scale, and recording all subjects' data. Each week that data were collected (the weeks in which all necessary staff were available), 12 patients were recruited into the study, half receiving intradermal buffered lidocaine and half bacteriostatic normal saline. A nurse researcher (SDB) collected participant data sheets and pharmacy spreadsheets weekly. A trained research assistant entered results into SPSS, version 16.0, statistical software.
RESULTS
The final sample (N = 148) was 65% women and 82% white (see Table 1). Only 4% and 1% of the study sample were known to be black and Hispanic/Latino, respectively, and 18 (12%) of the participants didn't provide race or ethnicity data. Participants ranged in age from 19 to 80, with a mean age of 52. A [chi]2 analysis revealed no significant differences in demographic characteristics between the two treatment groups. Of the final sample of 148 subjects, valid pain ratings were recorded for 145, of whom 73 received intradermal buffered lidocaine and 72 received intradermal bacteriostatic normal saline.
Reported perception of pain ranged from 0 to 9 in the buffered lidocaine group and from 0 to 10 in the bacteriostatic normal saline group. The mean pain rating was 1.56 (SD = 1.99) for intradermal buffered lidocaine and 2.58 (SD = 2.50) for intradermal bacteriostatic normal saline (see Table 2). Thus, buffered lidocaine was statistically significantly superior to bacteriostatic normal saline in reducing the pain of IV catheterization (t143 = -2.727; P = 0.007 [95% confidence interval, -1.77 to -0.28]).
Twice as many subjects who received intradermal buffered lidocaine as received intradermal bacteriostatic normal saline reported no pain at all upon IV insertion (30 [41.1%] versus 15 [20.8%], respectively) (see Figure 1). Only nine subjects (12.3%) receiving intradermal buffered lidocaine rated their pain at a level 5 or greater (the highest levels of pain reported), compared with 18 (25%) subjects receiving intradermal bacteriostatic normal saline. Overall, intradermal buffered lidocaine was consistently associated with less pain.
Various IV catheter sizes and insertion sites were used in this study, with 18- and 20-gauge over-the-needle catheters used in 53% and 45% of all cases, respectively. Rarely, a 16- or 22-gauge catheter was used. The protocol nurse was given no limitations related to IV insertion site selection, but in both treatment groups, the site used most commonly was the right forearm (32.4%), followed by the right hand (22.3%), left forearm (20.3%), left hand (12.8%), right wrist (7.4%), left wrist (3.4%), and right antecubital site (1.4%).
Distribution of subjects' sex and race or ethnicity, as well as catheter size and site used for venipuncture, was similar between the two treatment groups. Catheter size seemed to have little effect on pain; of the catheters used in this study, both the largest (16 gauge, n = 1) and smallest (22 gauge, n = 1) were used in the group pretreated with intradermal buffered lidocaine, which had the lowest mean pain rating. We also found no significant differences in pain rating according to age, sex, race or ethnicity, or location of IV site.
DISCUSSION AND NURSING IMPLICATIONS
The prevention and reduction of unnecessary pain is not only a universal nursing goal, it's now mandated by the Joint Commission. It's well documented that the discomfort associated with IV catheterization can be a cause of patient anxiety4 and dissatisfaction.23 The use of local anesthesia prior to IV catheterization reduces unnecessary pain and improves patient satisfaction, and has been shown to have no significant effect on either the time required for IV catheterization or its success rate.5,8 Previous research, however, hasn't provided clear guidance on selecting the most effective local anesthetic.
In contrast to the findings from Dava Brown's smaller study,1 the results of the current study indicate that adult patients undergoing same-day surgery experienced less pain when intradermal buffered lidocaine, rather than bacteriostatic normal saline, was used as a local anesthetic before IV catheterization. The anesthetic property of intradermal buffered lidocaine was more consistently effective than that of bacteriostatic normal saline; the mean pain score was lower and the standard deviation smaller in the buffered lidocaine group than in the bacteriostatic normal saline group.
The difference in findings between our study and Dava Brown's is probably a result of our larger sample size, but also may be related to the instrument used to capture pain assessment data (VAS in Brown's versus VNRS in our study). Brown's data also showed higher pain ratings when the IV was placed in the forearm. Our findings show that neither catheter size nor catheter site had any appreciable effect on pain perception with either type of solution.
Although statistical significance is of the utmost concern in evaluating research, from a clinical practice perspective nurses must balance statistical significance with clinical meaningfulness. Nursing practice must be guided by the best available evidence combined with clinical judgment, patient preferences, and available resources. For example, even though our study results indicate that intradermal buffered lidocaine is superior to intradermal bacteriostatic normal saline as an anesthetic pretreatment for IV catheterization, buffered lidocaine-which must be prepared in the pharmacy and replaced every seven to nine days11-requires more effort to use. The minimal cost difference between the two solutions fails to take into account the time required for the pharmacist to compound the lidocaine, the pharmacy oversight needed to ensure timely stock replacement, and the waste when medication expires and is discarded. In the same-day surgery environment, in which multiple IV lines are initiated daily, we feel that enough buffered lidocaine is used to justify the greater pharmacy effort and the waste. On a clinical unit in which fewer IV lines are started, dwell time is longer, or pharmacy staff is more limited, clinicians may determine that the need and benefit aren't great enough to justify such costs.
Potential allergy. According to our research, the best standard of care prior to IV catheterization is pretreatment with intradermal buffered lidocaine unless it's unavailable or the patient reports a "caine" allergy. Wilson and Martin addressed the issue of potential lidocaine allergy in their study of local anesthetics.14 Citing anecdotal accounts dating back 40 years, they noted that reports of anaphylactic reactions are rare and suggested that the methylparaben used to preserve lidocaine was the most likely cause of an allergic response. In patients who are allergic to lidocaine, bacteriostatic normal saline, which also has an analgesic effect, is an acceptable alternative.
Institutional policy change. Our literature review increased awareness throughout the hospital of the overwhelming evidence supporting the use of local anesthesia prior to IV catheterization. Our medical center responded by amending institutional policy on local anesthesia for IV catheterization, making it an expectation rather than a choice. Until that point, no nursing unit other than same-day surgery had made anesthetic pretreatment for venipuncture a routine practice. The new hospital policy also states that intradermal buffered lidocaine is preferred over intradermal bacteriostatic normal saline as anesthetic pretreatment for venipuncture, except in the case of allergy.
Policy change, of course, doesn't automatically result in a change in practice. The staff nurses in our hospital who conducted the research were ready to use intradermal buffered lidocaine consistently as a pretreatment for venipuncture, but the research team realized that the degree of acceptance of this change was likely to vary throughout the institution. To encourage practice change, research findings were shared with nursing staff at Nursing Practice Council and Clinical Ladder meetings. Presentations were made at leadership meetings, and after gaining leadership support, researchers communicated findings to nursing unit managers and staff throughout the hospital, one nursing unit at a time. As staff were offered education and provided support, practice changed. Follow-up continues through Nursing Practice Council meetings.
LIMITATIONS
Study limitations include the fact that an overwhelming majority of our subjects were white. Given the demographics of the community, the homogeneity of the population was not unexpected, but it does limit the generalizability of the findings to patients of different races, cultures, and ethnicities. Moreover, since this study was conducted on a same-day surgery unit in which all subjects were scheduled for IV catheterization, preoperative teaching prior to admission might have reduced anxiety, thereby producing different results from those seen in patients admitted to different types of nursing units. Finally, our research was specific to adults and cannot be generalized to children.
CONCLUSIONS
Patients rely on nurses to provide pain relief. By using local anesthesia prior to IV insertion, nurses may lessen the pain associated with this procedure. Our findings indicate that intradermal buffered lidocaine is superior to bacteriostatic normal saline as an anesthetic pretreatment for venipuncture; however, this study demonstrates that intradermal bacteriostatic normal saline can provide reasonable pain relief and is thus a practical alternative when intradermal buffered lidocaine is not available or the patient has an allergy to lidocaine.
REFERENCES