Regular monitoring of blood glucose levels is an important component
of routine diabetes care.1 Capillary blood is typically sampled
with the use of a fingerstick device and tested with a portable glucometer.
Because of outbreaks of hepatitis B virus (HBV) infections associated with
glucose monitoring, CDC and the Food and Drug Administration (FDA) have recommended
since 1990 that fingerstick devices be restricted to individual use.2,3 This report describes three recent outbreaks of HBV infection among
residents in long-term–care (LTC) facilities that were attributed to
shared devices and other breaks in infection-control practices related to
blood glucose monitoring. Findings from these investigations and previous
reports suggest that recommendations concerning standard precautions and the
reuse of fingerstick devices have not been adhered to or enforced consistently
in LTC settings.2-5 The findings underscore the need for education,
training, adherence to standard precautions, and specific infection-control
recommendations targeting diabetes-care procedures in LTC settings4-6 (see Box 1 ).
The three outbreaks described in this report were all reported by state
or local health departments to CDC, which provided epidemiologic and laboratory
assistance. In each of the three LTC settings, residents were tested for serologic
markers for HBV infection. Under the case definitions used in these investigations,
residents who tested positive for IgM antibody to hepatitis B core antigen
(anti-HBc) were defined as having acute HBV infection. Residents who tested
positive for hepatitis B surface antigen (HBsAg) and total anti-HBc, but who
tested negative for IgM anti-HBc, were considered to have chronic HBV infection.
Residents who tested positive for total anti-HBc, but who tested negative
for HBsAg, or those who had antibody to HBsAg (anti-HBs) ≥10 milli-International
Units (mIU) per milliliter were considered immune to HBV infection. Residents
were considered susceptible to HBV if they had no HBV markers. A retrospective
cohort study was performed as part of each investigation; the study was restricted
to acutely infected and susceptible residents to identify risk factors. In
all three investigations, staff members were evaluated; none were identified
as sources of infection. Medical records were reviewed and infection-control
procedures were assessed through direct observation and by interviews with
nursing staff members.
During November-December 2003, the Mississippi Department of Health
received reports of two fatal cases of acute HBV infection among residents
of nursing home A. The first patient with recognized symptoms of HBV infection
had received serologic testing for viral hepatitis infection in June 2003
as part of a hospital emergency department evaluation for abdominal pain.
Although this patient was found to have a positive test for IgM anti-HBc,
indicating acute HBV infection, and the finding was noted in the patient’s
chart in September 2003, nursing home A did not contact the state health department
or initiate an internal investigation. Subsequently, the patient died.
In December 2003, after a second patient with acute HBV infection had
died, and after a third with acute HBV infection was reported, serologic testing
was performed on specimens from all 158 residents. Test results were available
for 160 residents, including the two decedents; 15 (9%) had acute HBV infection,
one was chronically infected, 15 (9%) were immune, and 129 (81%) were susceptible.
Percutaneous and other possible exposures among residents were evaluated.
Among 38 residents who routinely received fingersticks for glucose monitoring,
14 had acute HBV infection, compared with one of 106 residents who did not
receive fingersticks (relative risk [RR] = 39.0; 95% confidence interval [CI]
Glucose monitoring of 14 residents with acute HBV infection and the
resident with chronic HBV infection was performed by staff members based at
the same nursing station. Reviews of infection-control practices and site
inspections indicated that each of the four nursing stations in nursing home
A was equipped with one glucometer and one spring-loaded, pen-like fingerstick
device. Staff members reported that a new end cap and lancet assembly was
used for each fingerstick procedure; however, the spring-loaded barrel and
glucometer were not routinely cleaned between patients. Investigators also
observed that insulin and other multidose medication vials were not labeled
with patient names or the dates the vials were opened. In an anonymous survey,
several staff members reported observing other workers reuse a needle or lancet
or fail to change gloves between patients. No other percutaneous exposures
were associated with illness.
During January-February 2004, the Los Angeles County Department of Health
Services received reports of four residents with diabetes in assisted living
center B who had acute HBV infection during November 2003–January 2004.
Because these initial reports were among residents with diabetes, serologic
testing was performed in January 2004 on residents who had received fingersticks
for blood glucose monitoring during May-December 2003. Of 22 residents tested
(three declined), eight (36%) had acute HBV infection, including the four
residents previously identified; six (27%) were immune (and excluded from
the analysis), and none had chronic infection. Reviews of patient records
indicated that one of the acutely infected residents had been repeatedly tested
at a separate hemodialysis center and had seroconverted to HBsAg-positive
in July 2003. Of the nine patients who had daily exposure to fingerstick procedures
performed by nursing staff, eight had acute HBV infection, compared with none
among the seven residents who performed their own fingersticks (RR = undefined;
CI = 2.8–undefined). Although receipt of insulin was also significantly
associated with infection, two residents with acute HBV infection had not
received insulin. Other percutaneous exposures (e.g., podiatric or dental
care) were not associated with HBV infection.
Fingerstick procedures were often performed by nursing staff members
in a central living area, with diabetes patients seated at a common table.
Although residents had their own fingerstick devices, nurses reported occasionally
using a pen-like fingerstick device barrel from their own kits to collect
consecutive blood samples; a single glucometer was typically used for all
residents. Nurses reported that they were discouraged from wearing gloves
to decrease the sense of a clinical environment, and hand hygiene was not
performed between procedures.
In May 2003, a case of HBV infection in a resident of nursing home C
was reported to the North Carolina Department of Health. During June-July
2003, serologic testing was performed on specimens from all 192 residents;
11 (6%) had acute HBV infection, 16 (8%) were immune, and 165 (86%) were susceptible.
No resident had chronic HBV infection. Of 45 residents who received fingersticks
for glucose monitoring, eight(18%) had acute HBV infection, compared with
three (3%) of 117 residents without this exposure (RR = 6.9; CI = 1.9-25.0).
After data were controlled for fingerstick exposures, acute HBV infection
was not associated with other percutaneous exposures (e.g., insulin injections,
podiatry procedures, or phlebotomy). Two diabetes patients at nursing home
C who were potential sources of the outbreak were identified retrospectively;
one had clinical symptoms of hepatitis B and serologic markers of acute infection
during 2002, whereas the other had chronic HBV infection and died in February
Interviews with staff and direct observation of glucose-monitoring practices
revealed that only single-use lancets were used, and insulin vials were not
shared among patients. However, on each wing of the facility, a single glucometer
was used for all patients receiving fingersticks; glucometers were not routinely
cleaned between patients. On some days, a single health-care worker performed
approximately 20 fingerstick procedures during a single work shift. In an
anonymous survey, nursing staff members indicated that some health-care workers
did not always change gloves between patients when performing fingerstick
R Webb, MD, M Currier, MD, J Weir, KM McNeill, MD, Mississippi Dept
of Health. E Bancroft, MD, D Dassey, MD, J Maynard, D Terashita, MD, Los Angeles
County Dept of Health Svcs, California. K Simeonsson, MD, A Chelminski, J
Engel, MD, North Carolina Dept of Health and Human Svcs. JF Perz, DrPH, AE
Fiore, MD, IT Williams, PhD, BP Bell, MD, Div of Viral Hepatitis, National
Center for Infectious Diseases; T Harrington, MD, C Wheeler, MD, EIS officers,
Lack of adherence to standard precautions and failure to implement long-standing
recommendations against sharing fingerstick devices place LTC residents at
risk for acquiring infections from bloodborne pathogens such as HBV.2,3,7 In nursing home A, the spring-loaded barrel of a fingerstick device
was used for multiple patients. Previous outbreaks have been linked to such
devices when the platform or barrel supporting the disposable lancet was reused
for multiple patients, when used lancets were stored with unused lancets,
or when lancet caps were reused (CDC, unpublished data, 1999).2,3 In
assisted living center B, nursing staff members routinely administered fingersticks
without wearing gloves or performing hand hygiene between patients, and spring-loaded
fingerstick devices were also occasionally shared.
In nursing home C, as with other recent outbreaks (CDC, unpublished
data, 2002),8 transmission of HBV among residents with diabetes
occurred despite use of single-use fingerstick devices or insulin medication
vials that were dedicated for individual patient use. In these settings, glucose
monitors, insulin vials, or other surfaces contaminated with blood from an
HBV-infected person might have resulted in transfer of infectious virus to
a health-care worker’s gloves and to the fingerstick wound or subcutaneous
injection site of a susceptible resident. Similar indirect transmission of
HBV in health-care settings through contaminated environmental surfaces or
inadequately disinfected equipment has been reported with other health-care
procedures, such as dialysis.6,9 HBV is stable at ambient temperatures;
infected patients, who often lack clinical symptoms of hepatitis, can have
high concentrations of HBV in their blood or body fluids.6 To prevent
patient-to-patient transmission of infections through cross-contamination,
health-care providers should avoid carrying supplies from resident to resident
and avoid sharing devices, including glucometers, among residents.
The risk for patient-to-patient transmission of HBV infection can be
reduced by implementing specific prevention measures (see Boxes 1 and 2 ). LTC staff often perform
numerous percutaneous procedures; frequent blood glucose monitoring increases
opportunities for bloodborne pathogen transmission. The outbreak investigations
reported here identified residents with diabetes who received fingersticks
from nursing staff members as often as four times per day, according to their
physician’s routine orders, despite having consistently normal glucose
levels. Expert panels have concluded that approximately 8 years are needed
before the benefits of glycemic control result in reductions in microvascular
complications.1,10 In LTC settings, schedules for fingerstick blood
sampling of individual patients should be reviewed regularly to reduce the
number of percutaneous procedures to the minimum necessary for their appropriate
medical management. In each of the investigations described in this report,
implementation of infection-control measures (see Boxes 1 and
2 ) was recommended, along with follow-up serologic
testing for markers of HBV.
An estimated 70,000-80,000 HBV infections occur each year in the United
States. Most of these infections occur among young adults with behavioral
risk factors (i.e., sexual contact and injection-drug use); these adults should
receive hepatitis B vaccine. Preventing transmission of HBV among patients
in long-term–care settings requires adherence to recommended infection-control
practices and prompt response to identified instances of transmission. Routine
hepatitis B vaccination or screening of LTC residents is not recommended.
In the outbreaks described in this report, initial cases were not identified
or investigated in a timely fashion, resulting in missed opportunities to
correct deficient practices and interrupt transmission. Evidence of acute
viral hepatitis in any LTC resident should prompt a thorough investigation.
For a case involving a resident with diabetes, fingerstick blood sampling
procedures and insulin administration should receive particular scrutiny.
Health departments should encourage reporting of such cases and offer assistance
in identifying the source of infection. CDC continues to support investigations
in LTC and other health-care settings and is working toward improved implementation
of the infection-control recommendations described in this report.
The findings in this report are based, in part, on data provided by
C Ranck, R Hotchkiss, B Amy, MD, Mississippi Dept of Health. J Rosenberg,
MD, Div of Communicable Disease Control, California Dept of Health Svcs. P
MacDonald, PhD, Dept of Epidemiology, Univ of North Carolina, Chapel Hill;
S Smith, P Poole, North Carolina Dept of Health and Human Svcs. M Viray, Epidemiology
Program Office, CDC.
REFERENCES: 10 available
Transmission of Hepatitis B Virus Among Persons Undergoing Blood Glucose
Monitoring in Long-Term–Care Facilities—Mississippi, North Carolina,
and Los Angeles County, California, 2003-2004. JAMA. 2005;294(1):35–38. doi:10.1001/jama.294.1.35
* * SCHEDULED MAINTENANCE * *
The JAMA Network Sites will be conducting routine maintenance from 10/20/2017 through 10/21/2017. During this window access to content and authentication may be intermittently available. The JAMA Store will be completely unavailable during the maintenance window.