Article
Editorial Note
References
POINT OF CONTACT FOR THIS
DOCUMENT:
Tables
Evaluation of three safety devices used in phlebotomies based ...
Health-care
workers (HCWs) are at risk for infections with bloodborne pathogens resulting
from occupational exposures to blood through percutaneous injuries (PIs). Phlebotomy,
one of the most commonly performed medical procedures, has been associated with
13%-62% of injuries reported to hospital occupational health services (1,2) and
with 20 (39%) of the 51 documented episodes of occupationally acquired human
immunodeficiency virus (HIV) infection reported in the United States (CDC,
unpublished data, 1996). Although safety devices designed to prevent PIs
associated with phlebotomy have been available for use in the United States,
clinical evaluation of these devices has been difficult because 1)
ascertainment of PIs is difficult (many injuries are unreported {2,3}, and
observation of all procedures is impractical because phlebotomy is performed
throughout the hospital by different groups of HCWs at all hours), 2) data to
calculate PI rates (i.e., the number of phlebotomies performed and devices
used) are not routinely available, 3) a large number of phlebotomies must be
evaluated because of the low rates of phlebotomy-related PI, and 4) rates of
safety-feature activation are difficult to assess. This report summarizes a
collaborative study by CDC and six hospitals to evaluate safety devices for
phlebotomy. The findings indicate that use of safety devices significantly
reduced phlebotomy-related PI rates while having minimal clinically apparent
adverse effects on patient care. *
The study was conducted in
two phases during 1993-1995 at six university-affiliated hospitals in
Minneapolis-St. Paul, Minnesota (three hospitals), New York, New York (one
hospital), and San Francisco, California (two hospitals). Each hospital
selected the products to be evaluated (vacuum-tube blood-collection devices
and/or winged steel needles with safety features). The assessment was
restricted to a comparison of safety devices with conventional devices, not
with other safety devices. Products evaluated included a resheathable winged
steel needle (Safety-Lok{trademark} {Becton Dickinson, Franklin Lakes, New
Jersey} ** {six hospitals}); a bluntable vacuum-tube blood-collection needle
activated while in the patient's vein (Punctur-Guard{trademark} {Bio-Plexus,
Inc., Tolland, Connecticut} {three hospitals}); and a vacuum-tube
blood-collection needle with a hinged recapping sheath (Venipuncture
Needle-Pro{trademark} {Smith Industries (Concord Portex), Keene, New Hampshire}
{four hospitals}). Each product requires the HCW to activate the safety feature
during or after phlebotomy. Before introducing safety devices, each hospital
conducted a comprehensive training program for HCWs that included
"hands-on" experience with the equipment.
During phase I (mean
duration among the hospitals: 10 months; range: 9-12 months), hospitals used
conventional phlebotomy devices and conducted enhanced surveillance for
injuries (e.g., encouraging reporting, publishing notices in the hospital
newsletter, posting educational materials, and/or providing inservice training
for staff). An anonymous survey was distributed to four groups of HCWs who
routinely perform phlebotomies *** to estimate their rates of underreporting of
PIs to hospital surveillance systems and to determine the average number of
phlebotomies performed each day and average number of days worked each week. The
rates of PIs associated with phlebotomy devices for HCWs in each of these four
groups were estimated by dividing the number of phlebotomy-related PIs reported
to the hospital's surveillance system during the study period (adjusted for
underreporting by occupation) by the total number of phlebotomies performed
(estimated from the daily average number of phlebotomies performed by each HCW,
the number of HCWs in each of the four groups, and the duration of the study
period).
During phase II (mean
duration among the hospitals: 12 months; range: 6-15 months), investigators
replaced conventional phlebotomy devices with safety devices hospitalwide,
monitored supplies of phlebotomy equipment to attempt to ensure that only
safety devices were available, continued enhanced surveillance for injuries,
and inventoried the autoclaved contents of a representative sample of disposal
containers for sharp instruments to determine rates of use of safety devices
and conventional devices and rates of activation of safety features. The HCW
survey was repeated 1-2 months before the end of phase II, and the estimated PI
rates for safety and conventional devices were compared. The second HCW survey
also included questions to assess HCW satisfaction with safety devices and to
determine the occurrence of adverse effects in patients that were apparent at
the time of the phlebotomy ****.
The overall response rate
for each of the two HCW surveys was approximately 75%, based on estimates of
the number of HCWs who received survey forms; 1699 HCWs responded in phase I
and 1421 in phase II. Overall, respondents acknowledged reporting 302 (54%) of
563 needlestick injuries they had sustained from all types of needles during
the previous year. Reporting rates varied by occupation: 91% of injuries among
phlebotomists were reported, as were 68% among nurses, 35% among medical
students, and 31% among residents. Within occupations, reporting rates were
similar among hospitals and between the two surveys. Because estimated rates of
phlebotomy-related PI by device and occupation were similar for each hospital
in which a particular device was used, data were aggregated among hospitals to
permit comparison of PI rates for safety and conventional phlebotomy devices. Compared
with conventional devices, PI rates were lower for safety devices (Table 1).
Of 41 PIs associated with
safety devices, 34 (83%) involved winged steel needles and seven (17%) involved
vacuum-tube blood-collection needles (Table 1). Twenty-five (61%) involved an
injury before activation of the safety feature was appropriate or possible
(e.g., within seconds after the device was removed from the vein); six (15%)
occurred during activation of the safety feature (all with
Safety-Lok{trademark}). For eight (20%), the safety feature had not been
activated, and for two (5%), the mechanism of injury was unknown. Safety
devices constituted 12,681 (89%) of the 14,261 phlebotomy devices in autoclaved
sharps-disposal containers. In the phase II HCW survey, HCWs were asked
"Do you prefer the safety device over conventional equipment?" Among
1108 HCWs, 1879 responses were related to one or more of the three devices; 822
(44%) responded yes; 622 (33%), no; and 435 (23%), unsure.
Reported by: M Mendelson,
MD, R Solomon, MS, E Shekletski, Mt. Sinai Medical Center, New York City. K
Henry, MD, S Campbell, MSPH, A Collins, St. Paul Ramsey Medical Center, St.
Paul; J Thurn, MD, F Lebahn, MT, Minneapolis Veterans Affairs Medical Center,
Minnesota; F Rhame, MD, Univ of Minnesota Hospital, Minneapolis. J Gerberding,
MD, R Fahrner, MS, K Turner-Hubbard, MS, San Francisco General Hospital; P
Jensen, MD, San Francisco Veterans Affairs Medical Center, California. Hospital
Infections Program, National Center for Infectious Diseases, CDC.
Editorial
Note: The findings in this report suggest that safety devices for phlebotomy
can reduce the risk for occupational PIs among HCWs. In particular, there was a
significant reduction in phlebotomy-related PIs associated with use of each of
the vacuum-tube blood-collection devices and a reduction in PIs associated with
use of the winged steel needles. Further decreases in phlebotomy-related PIs
might have been possible with increased use of safety devices and/or increased
activation of safety features by HCWs. Experts have recommended that safety
devices include safety features that activate automatically and do not rely on
activation by HCWs (4,5). Although the assessment of potential patient
complications in this study was limited, short-term complications were
clinically minimal, and although patients were not systematically monitored for
long-term follow-up, phlebotomy needles are not indwelling devices and
long-term complications of phlebotomy are rare.
Results of this study also
suggest that safety devices for phlebotomy may be generally acceptable to
users. Activation rates of safety features and user acceptability may be
influenced by factors such as the perceived risk for occupational infection by
the HCW, design of the device, training provided before and after introduction
of the device, length of time needed to become adept at using the device, ease
of use, necessary changes in technique, and previous experience with safety
devices (5). Further analyses will assess whether safety-feature activation
rates and user acceptability in this study varied by hospital, city,
occupation, or device used. Acceptability of a device to an institution may be
influenced by cost.
In this study, only 54% of
PIs were reported to hospital surveillance systems -- a rate consistent with
those documented in previous studies (range: 5%-60% {2,3}). Failure to report
PIs may compromise appropriate postexposure management, including postexposure
prophylaxis for HIV and hepatitis B virus, and assessment of occupational
hazards and preventive interventions (6,7). Health-care institutions and HCWs
must further assess reasons for underreporting and improve reporting of all
occupational blood exposures.
The Occupational Safety and
Health Administration requires that primary methods to reduce occupational PIs
include engineering controls (8), and the Food and Drug Administration has
urged that needleless or recessed needle systems be used to replace hypodermic
needles for accessing intravenous administration sets (9). Some manufacturers
are continuing efforts to develop and refine safety devices to improve the
effectiveness and acceptability of products. The findings in this report and in
a companion report evaluating blunt suture needles (10) suggest that safety
devices can be an effective component in a needlestick-prevention program. The
Public Health Service is evaluating the implications of these and other data in
assessing the possible need for further guidance on selection, implementation,
and evaluation of safety devices in health-care settings.
References
* Single
copies of this report will be available free until January 16, 1998, from the
CDC National AIDS Clearinghouse, P.O. Box 6003, Rockville, MD 20849-6003;
telephone (800) 458-5231 or (301) 217-0023.
** Use of trade names and
commercial sources is for identification only and does not imply endorsement by
the Public Health Service or the U.S. Department of Health and Human Services.
*** Phlebotomists
(including laboratory technicians who frequently draw blood); nurses (on
representative medical and surgical wards, intensive-care units, and in the
emergency department); residents (medical, pediatric, and surgical); and
medical students (third- and fourth-year).
**** Examples of adverse
effects include vein trauma resulting in hematoma, increased patient
discomfort, and the need for repeated phlebotomy attempts. Because certain
events reported as patient adverse effects (e.g., slow blood return or other
difficulty drawing blood, sometimes requiring repeat phlebotomies) also were
considered technical difficulties, responses were classified as "adverse
patient effects or technical difficulties."
To request
a copy of this document or for questions concerning this document, please
contact the person or office listed below. If requesting a document, please
specify the complete name of the document as well as the address to which you
would like it mailed. Note that if a name is listed with the address below, you
may wish to contact this person via CDC WONDER/PC e-mail.
For single issue purchase 800-843-6356
HOSPITAL INFECTIONS PROGRAM
State/Fed Gov: For free copies
write to: CDC, MMWR MS(C-08)
Atlanta, GA 30333
TABLE 1. Evaluation of three safety devices * used in phlebotomies based on surveillance and surveys of health-care workers (HCWs) +, by
characteristic -- Minneapolis-St. Paul, New York City, and San Francisco, 1993-1995 &
==============================================================================================================================================================================
Vacuum-tube blood-collection device
------------------------------------------------------------
Characteristics Winged steel needle Safety-Lok @ Punctur-Guard ** Venipuncture Needle-Pro
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Study site (no. hospitals) Minneapolis- St. Paul (3) Minneapolis- St. Paul (3) Minneapolis- St. Paul (1)
New York City (1) New York City (1)
San Francisco (2) San Francisco (2)
No. phlebotomy-related percutaneous injuries (PIs)
Unadjusted
Conventional device 53 14 19
Safety device 34 2 5
Adjusted for underreporting by occupation
Conventional device 102 19 33
Safety device 58 4 8
Estimated no. phlebotomies performed
Conventional device 2,540,500 523,561 895,054
Safety device 1,875,995 501,596 628,092
Estimated no. PIs per 100,000 phlebotomies
Conventional device 4.0 3.6 3.6
Safety device ++ 3.1 0.9 1.2
Percentage reduction in PI rate with safety 23% (p=0.07) 76% (p=0.003) 66% (p=0.003)
device++
No. (%) safety devices with activated safety 2257 (56%) of 4065 2984 (57%) of 5255 3250 (98%) of 3319
features observed in disposal containers
No. (%) HCWs noting technical difficulties or 97 (10%) of 955 201 (44%) of 452 19 (5%) of 385
adverse patient effects with safety device
(preliminary results) &&
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
* Safety-Lok (Becton Dickinson, Franklin Lakes, New Jersey), Punctur-Guard (Bio-Plexus, Inc., Tolland, Connecticut), and Venipuncture Needle-Pro (Smith Industries {Concord
Portex}, Keene, New Hampshire.
+ Phlebotomists; nurses on representative medical and surgical wards, intensive-care units, and the emergency department; medical, pediatric, and surgical residents; and
third- and fourth-year medical students.
& This study was not designed to compare one safety device with another.
@ Use of trade names and commercial sources is for identification only and does not imply endorsement by the Public Health Service of the U.S. Department of Health and
Human Services.
** According to the manufacturer, the design of this product has been modified since study completion.
++ Safety versus conventional device.
&& Approximately 60% of respondents answered this question. Perception of technical difficulties may be influenced by training provided, length of time using the
device, perception of risk for occupational infection, and other factors.
==============================================================================================================================================================================