Share this article on:

A Prospective Randomized Trial of Two Safety Peripheral Intravenous Catheters

Prunet, Bertrand MD; Meaudre, Eric MD; Montcriol, Ambroise MD; Asencio, Yves MD; Bordes, Julien MD; Lacroix, Guillaume MD; Kaiser, Eric MD

doi: 10.1213/ane.0b013e318174df5f
Patient Safety: Research Report

BACKGROUND: To reduce the risk of accidental needlestick injuries, first active then passive safety devices were developed on IV catheters. However, whether these catheters are easy to implement and really protect personnel from accidental needlestick is untested.

METHODS: In this prospective randomized survey, we compared a passive safety catheter with an active safety catheter and a nonsafety classic catheter. The main objective was to evaluate the difficulty of inserting the catheters in terms of the number of insertion failures, difficulties introducing the catheter and withdrawing the needle, and the normality of the blood reflux in the delivery system. The second objective was to determine the degree of exposure to patients’ blood evaluated as the number of exposures of the staff and blood splashes of the environment, and the staff’s sense of protection.

RESULTS: Seven hundred fifty-nine assessment cards were collected. The number of failures for the three catheter groups was similar and not statistically different. Introduction of the catheter was more difficult with the active safety catheter. Needle withdrawal was more difficult with the passive safety catheter. The blood reflux was abnormal more often with the safety catheters. The staff’s exposure was more frequent with the active safety catheter. The number of blood splashes was more common with the safety catheters.

CONCLUSIONS: Safety catheters are not superior with regard to failure rate in the catheter’s placement. Users feel better protected, but find the use of safety catheters more difficult, and their handling generates more splashing of blood into the environment. The passive safety catheter is more efficient than the active safety catheter with regard to ease of introduction of the catheter into the vein and the staff’s exposure to the patient’s blood.

IMPLICATIONS: Whether the security system is active or passive, safety IV catheters do not have a greater incidence of placement failure than standard catheters. Users feel better protected, but find their use more difficult, and their handling generates more splashing of blood into the environment.

From the Department of Anesthesiology, Military Teaching Hospital Sainte Anne, 83800 Toulon Armées, France.

Accepted for publication March 10, 2008.

Supported by Sainte Anne Hospital and Department of Anesthesiology.

Address correspondence and reprint requests to Bertrand Prunet, Département d’Anesthésie, Hôpital d’Instruction des Armées Sainte Anne, 83800 Toulon Armées, France. Address e-mail to prunet.bertrand@orange.fr.

Accidental needlestick injuries expose health-care workers to blood-transmitted infectious agents. Experience shows that even an intense educational policy with good organization of the care, staffing, and safety equipment is insufficient to prevent accidental needlesticks.1–3 Staff in the anesthesia and emergency department reception areas are especially prone to exposure.4–6 To reduce the risk of accidental needlestick with IV catheters, active safety devices were developed during the 1990s.6 The operator triggers the device using a specific maneuver. In the 2000s, passive safety devices have appeared on the market, which do not require any particular maneuver from the operator because they are automatically triggered.7 These new types of passive safety IV catheters are placed according to a technique identical to the classic nonsafety catheters, but no randomized prospective survey has ever assessed them. The question is whether their insertion is easy and whether they really protect personnel from accidental needlestick. The goal of this prospective randomized survey was to compare a recently marketed passive security catheter (Introcan® Safety™) with an active security catheter (Insyte™ Autoguard™) and a nonsafety classic catheter (Vialon™). The main objective of the survey was to evaluate the difficulty of inserting the catheters. The second objective was to determine the degree of exposure of personnel to patients’ blood.

Back to Top | Article Outline

METHODS

After agreement of the local ethics committee, a randomized prospective survey was conducted in the operating room and the emergency department over a 5 Mo period to assess 3 peripheral IV 18-gauge diameter catheters. Vialon (BD Medical Systems, Le Pont de Claix, France) was the nonsafety catheter usually used in the hospital (18 G, length 30 mm, outer diameter 1.3 mm). Passive safety Introcan Safety catheters (B.Braun Médical, Boulogne, France) have an insertion identical to that of Vialon with a protective shield that automatically covers the needlepoint during its withdrawal from the catheter top without any specific intervention from the operator (18 G, length 32 mm, outer diameter 1.3 mm). Active safety Insyte Autoguard (BD Medical Systems, Le Pont de Claix, France) requires pressing a button to trigger the withdrawal of the needle in a plastic sleeve using a spring (18 G, length 30 mm, outer diameter 1.3 mm). The length of the survey was previously fixed as the time necessary to obtain at least 250 informed assessment cards for every type of catheter. Immediately before every procedure, the type of peripheral venous catheter to use was determined randomly in a three-ball ballot box. The operators were the anesthetists–physicians and the anesthetists–nurses in the operating room, and the emergency physicians and the nurses in the emergency department. All personnel were educated on how to place the different catheters, thanks to specific posters provided by the manufacturers. To be qualified for the protocol, adult patients had to enter the operating room or the emergency department during the study period and require an IV infusion. All patients gave informed written consent to participate in this study. During the insertion of the catheter, after application of a venous tourniquet at the level of the arm and cutaneous disinfection, the operator assessed the quality of the patient’s peripheral veins at the level of the back of the patient’s hand or forearm. If the operator considered the patient’s veins unsuitable for placing an 18-G catheter, the patient was excluded from the protocol. If the operator considered the patient’s veins suitable for placing an 18-gauge catheter, the patient was included in the protocol and the operator attempted to place the catheter into the vein that appeared most adequate. The choice of the catheter was randomized by using a single-blinded envelope method. In case of failure on the first attempt, the operator made a second attempt, always with the same type of catheter. In case of failure on the second attempt, the insertion of this catheter was considered a definitive failure. Immediately after the procedure, the operator completed an assessment card. The difficulty in using the catheter was assessed by the number of insertion failures in the first attempt (Failure 1) and in the second attempt (Failure 2). The difficulty of introducing the catheter into the vein and withdrawing the needle was assessed by the operator immediately after procedure using a Visual Analog Scale (VAS), ranging from 0 (very easy) to 10 (very difficult). The blood reflux in the catheter delivery system was considered normal when the filling was progressive and complete in <4 S. It was considered abnormal when the filling was incomplete or complete but too slow (>4 S). The number of exposures of the staff to the patient’s blood was assessed by identifying the cases where the patient’s blood stained the operator’s skin, gloves, mask, or any other clothing. The number of blood splashes of the environment was assessed by identifying cases in which the patient’s blood stained the stretcher or floor. A VAS used by the operator, ranging from 0 (very safe catheter) to 10 (not at all safe catheter), was used to assess the staff’s sense of protection against the risk of an accidental needlestick immediately after the procedure.

The data are presented as numbers (%) for nominal variables, as the mean ± sd for continuous variables and as median [quartile 0.25–quartile 0.75] (mean) for the VAS scores. Comparison of the groups was made by Pearson’s Khi-two test for nominal variables and by the Mann-Whitney’s U-Test for continuous variables. A value of P < 0.05 was considered significant.

Back to Top | Article Outline

RESULTS

A total of 759 assessment cards were collected. Two hundred fifty-four cards were collected in the Vialon group, 251 cards were collected in the Introcan Safety group, and 254 cards in the Insyte Autoguard group. The patient characteristics were similar for the three IV catheters (Table 1). No accidental needlesticks occurred during the survey. The results of the variables assessing the difficulty of placing the catheter are presented in Table 2. The number of failures was similar and not statistically different for the three groups of catheters for the first and the second attempts. Introduction of the catheter into the vein was more difficult in the Insyte Autoguard group compared with the Vialon and the Introcan Safety groups. The withdrawal of the needle was more difficult in the Introcan Safety group compared with the Vialon and the Insyte Autoguard groups, and more difficult in the Insyte Autoguard group compared with the Vialon group. During the survey, no case of accidental withdrawal of the catheter due to the withdrawal of the needle was reported. The blood reflux in the catheter delivery system was more often abnormal with the Vialon group compared with the Introcan Safety and the Insyte Autoguard groups, and compared with the Introcan Safety group for the Insyte Autoguard group. The results of the variables assessing the exposure of personnel to the patients’ blood and the staff’s sense of protection are presented in Table 3. The staff’s exposure to the patients’ blood occurred more frequently in the Insyte Autoguard group than in the Vialon group and the Introcan Safety group. The number of blood splashes in the environment was greater in the Introcan Safety group and in the Insyte Autoguard group compared with the Vialon group. The staff’s sense of protection against the risk of an accidental needlestick was more favorable with the safety catheters than with the conventional catheter.

Table 1

Table 1

Table 2

Table 2

Table 3

Table 3

Back to Top | Article Outline

DISCUSSION

This prospective, randomized survey shows that if safety IV catheters are uncomfortable for the operator to use, they are not objectively more difficult to place than the nonsafety catheter. The high number of patients included is a considerable strength of this survey (more than 250 patients per group) in contrast to previous studies that included 40 to 50 patients per group.8,9 This makes the present results highly relevant and constitutes an important argument in the choice of the type of catheter used at a hospital level. The rate of placement failures of the safety catheters was super-imposable on that of the conventional catheter. This information is also found in the literature8,10 where failure rates are similar for Insyte Autoguard (0%–2%) and for another passive security catheter, Protective® Acuvance® (Johnson & Johnson Medical, TX) (0%–4%). Only the active safety catheter was more difficult to introduce into the vein. The passive safety catheter was not more difficult to introduce into the vein than the standard nonsafety catheter, probably because the procedure for both is identical. With Insyte Autoguard, the voluminous safety appliance complicates the puncture.8,11 Two previous prospective studies assessing Insyte Autoguard produced conflicting results for this variable. One study concluded that the catheter was more difficult to introduce compared with the conventional catheter,8 whereas the other one did not.10 The withdrawal of the needle was more difficult for safety catheters than for the nonsafety catheter. Among safety catheters, withdrawal was more difficult for Introcan Safety. The mandrin slides less easily when withdrawn with Introcan Safety because of the shield around the needle. This problem has few practical consequences since the catheter is, at that moment, already in the vein. A prospective, randomized survey assessing 10 types of catheters with 40 patients in each group also concluded that all safety catheters with a needle in situ caused a resistance during the withdrawal of the needle, with no difference in the failure rate compared with nonsafety catheters.9

The blood reflux in the catheter delivery system was more often abnormal for safety catheters compared with the nonsafety catheter, which was previously observed in other studies.8,10,12 Moreover, among safety catheters in the present study, this reflux was more often abnormal for Insyte Autoguard than for Introcan Safety. Here again, the passive safety catheter was more efficient than the active safety one. No accidental needlestick occurred during the survey.

No published randomized prospective survey is sufficiently statistically powered to prove that safety venous catheters reduce the risk of a needlestick. Yet, a retrospective cohort survey including about 6500 venous access attempts during emergency medical services in the field found that the incidence of accidental needlesticks decreased from 169 to 0 for 100,000 patients due to the change from a standard venous peripheral catheter to a safety catheter.6 Nevertheless, the staff’s exposure to the patients’ blood remains a problem. In the present survey, exposure occurred more frequently with Insyte Autoguard than with the conventional nonsafety catheter and Introcan Safety. This information has already been studied in a prospective randomized survey including 330 children to assess an active security catheter where the nursing staff’s skin was more often stained than with the conventional catheter (61 vs 16 cases).12 This has not been found with adult patients.8,10 In any case, the environment was stained more often when a safety catheter was used,8,10,12 as was found in the present survey. This may have been due to a training effect. However, the theoretical objective of the passive protective system is to not modify the insertion technique, thus theoretically cancelling a training effect. The training curves for the insertion of each catheter were not investigated in this survey. Nevertheless, the staff’s sense of protection against the risk of an accidental needlestick was better with the safety catheters than with the conventional catheter, and this was found in all previous studies as well.6,8,10,12 This should cause the staff to better accept these new devices.11

In conclusion, whether the safety system is active or passive, safety catheters do not have a great incidence of placement failure compared with standard catheters. Users feel better protected, but find their use more difficult and their handling generates more splashing of blood into the environment. The passive safety catheter is more efficient than the active safety catheter with regard to the ease of introducing the catheter into the vein and the staff’s exposure to the patient’s blood.

Back to Top | Article Outline

REFERENCES

1. Sohn SJ, Eagan J, Sepkowitz KA, Zuccotti G. Effect of implementing safety-engineered devices on percutaneous injury epidemiology. Infect Control Hosp Epidemiol 2004;25:536–42
2. Rogues AM, Verdun-Esquer C, Buisson-Valles I, Laville MF, Lashéras A, Sarrat A, Beaudelle H, Brochard P, Gachie JP. Impact of safety devices for preventing percutaneous injuries related to phlebotomy procedures in health care workers. Am J Infect Control 2004;32:441–4
3. Clarke SP, Rockett JL, Sloane DM, Aiken LH. Organizational climate, staffing, and safety equipment as predictors of needlestick injuries and near-misses in hospital nurses. Am J Infect Control 2002;30:207–16
4. Greene ES, Berry AJ, Jagger J, Hanley E, Arnold WP, Bailey MK, Brown M, Gramling-Babb P, Passannante AN, Seltzer JL, Southorn P, Van Clief M, Venezia R. Multicenter study of contaminated percutaneous injuries in anesthesia personnel. Anesthesiology 1998;89:1362–72
5. Greene ES, Berry AJ, Arnold WP, Jagger J. Percutaneous injuries in anesthesia personnel. Anesth Analg 1996;83:273–8
6. O’Connor RE, Krall SP, Megargel RE, Tan LE, Bouzoukis JK. Reducing the rate of paramedic needlesticks in emergency medical services: the role of self-capping intravenous catheters. Acad Emerg Med 1996;3:668–74
7. Trim JC. A review of needle-protective devices to prevent sharps injuries Br J Nurs 2004;13:144–53
8. Asai T, Hidaka I, Kawashima A, Miki T, Inada K, Kawashi S. Efficacy of catheter needles with safeguard mechanisms. Anaesthesia 2002;57:572–7
9. Kondo M, Kim SJ, Fujiwara Y, Iinuma A, Koji K, Irie Y, Tazumi K, Noguchi S, Monden M. Evaluation of ten intravenous catheters for operability and safety in the infusion room. Gan To Kagaku Ryoho 2004;31:2005–8
10. Asai T, Matsumoto S, Matsumoto H, Yamamoto K, Shingu K. Prevention of needle-stick injury: efficacy of a safeguarded intravenous cannula. Anaesthesia 1999;54:258–61
11. Rivers DL, Aday LA, Frankowski RF, Felknor S, White D, Nichols B. Predictors of nurses’ acceptance of an intravenous catheter safety device. Nurs Res 2003;52:249–55
12. Coté CJ, Roth AG, Wheeler M, Rahe C, Rae BR, Dsida RM, Przybylo HJ. Traditional versus new needle retractable i.v. catheters in children: are they really safer, and whom are they protecting? Anesth Analg 2003;96:387–91
© 2008 International Anesthesia Research Society