- Email: [email protected]
In-hospital needlesticks and other significant blood exposures to blood from patients with acquired immunodeficiency syndrome and lymphadenopathy syndrome
American
Journal
Volume
Frank S. Rhame, M.D., Minneapolis, Minnesota and the membership of the Association for Practitioners
There have (AIDS) in assess the Practitioners percent of more than
of infection
12 Number
2
Control April
1984
in infection Control
been no documented cases of acquired immunodeficiency syndrome personnel after a prospectively recognized in-hospital blood exposure. To frequency of such exposure the membership of the Association for in Infection Control was surveyed in early June 1983. Thirty-three the membership responded, accounting for 42% of U.S. hospitals with 250 beds. Respondents reported needlestick or other significant blood
exposures to blood from patients with AIDS occurring in 157 instances and to blood from patients with lymphadenopathy syndrome in 43 instances. Nineteen and twenty of the exposures, respectively, occurred before July 1982. There are two bases for believing that AIDS will not pose a substantial risk to hospital workers: the lack of demonstrated in-hospital AIDS transmission to date and the recognition that other viruses besides the hepatitis B virus-viruses that seem to have less potential for in-hospital transmission-are equally plausible models of AIDS transmission. (AM J INFECT CONTROL 12:69-75, 1984.)
Although there is still no published case of clearly established in-hospital transmission of acquired immunodeficiency syndrome (AIDS), there is great concern about potential hazards From the Infection Control Program, University of Minnesota Hospitals and Clinics, and the Infectious Disease Section, Department of Medicine and Department of Laboratory Medicine and Pathology, University of Minnesota. Reprint requests: Frank S. Rhame, M.D., ing, University of Minnesota Hospitals neapolis, MN 55455.
Box 421 Mayo and Clinics,
BuildMin-
to hospital workers from AIDS patients. A valid basis for this concern is the close correspondence between the groups at risk of acquiring AIDS and those at increased risk of acquiring hepatitis B. Hepatitis B is the major hazard associated with hospital work, and AIDS patients tend to be high-acuity patients with great potential for blood exposure. In the hospital, blood has been the only important vector of hepatitis B transmission, and reduction of exposure to blood and blood-containing materials 69
Amerlcan
70
Rhame
has been considered the prime method of preventing transmission of AIDS.’ Thus needlestick exposure is a logical first place to look for evidence of AIDS transmission in the hospital. Direct evidence bearing on needlestick transmission of AIDS is scant. Relative nontransmissibility of AIDS to hospital workers is suggested by one preliminary study.2 Eight personnel had needlestick blood exposure between August and December of 1982. As of May, 1983, careful evaluation, including studies of cell-mediated immunity, showed no sign of AIDS transmission. A needle puncture was reported by one of seven heterosexual women who were regular sexual partners of male AIDS patients,3 but the woman showed no evidence of having acquired AIDS. A dase of AIDS has been reported in a sanitation worker with no risk of exposure to AIDS.4 He worked in the South Bronx, an area with many addicts. No accidental needlestick was recognized, but such exposure may have occurred. Four cases of AIDS in hospital workers with no recognized AIDS risk factors have recently been reported.5 However, only one was suggestive of hospital transmission: the employee worked as a housekeeper in an ambulatory surgery patient area in a Baltimore hospital. He had frequent exposure to blood and had had a needlestick caused by a discarded needle from an unknown source about 15 months before the diagnosis of AIDS. His negative history of other AIDS risk factors was partially confirmed. Information about the other hospital workers was less complete, but one was from the Caribbean and the other two had some suggestion of unconceded exposure to AIDS. None had a recognized exposure to a patient with AIDS. The lack of a single case of AIDS clearly resulting from hospital transmission is encouraging, but proper interpretation of this observation is hampered by two main problems: the long incubation period of AIDS and lack of denominator data describing hospital exposure to AIDS. The incubation period of AIDS in adults is generally considered to be at least a year. In a cluster of AIDS cases occurring in Orange County, California,G three patients with specific sexual contacts with AIDS patients developed symptoms 9, 13, and 22 months, respectively,
INFECTION
Journal
oi
CONTROL
after the contacts. Among 15 adults with AIDS whose only AIDS risk factor was blood transfusion, the mean interval from transfusion to onset of symptoms was 18 months.7 With respect to the lack of denominator data, the study reported herein provides some help. It presents information reported in a survey of the membership of the Association for Practitioners in Infection Control (APIC) conducted in June 1983 on the number of in-hospital needlesticks and other significant blood exposures to blood from AIDS patients and patients with iymphadenopathy syndrome.
As part of an effort to determine the feasibility of a national registry of hospital exposure to AIDS, a survey was mailed on June 3 or 6, 1983, to the 5834 APIC members with addresses in the United States. The mailing contained a cover letter describing a proposal to form the APIC National Registry of Hospital Exposure to AIDS, an explanation of the registry organization, and a questionnaire (Fig. 1). Responses to the third column of the questionnaire table, which requested data on exposures to blood from patients in the AIDS risk category, “other high AIDS risk,” were irregular and are not reported in this article. For making compilations, the following criteria were used: the specialty of the APIC member was that provided on the APIC mailing list; the specialty designations were supplied by the applicant at the time of application to APIC, and some had been updated at the time of membership renewal. Hospital information was taken from the 1972 American Hospital Association Guide? Standard metropolitan statistical area boundaries were taken from City and County Data Book.!” -Ts By July 27, 1983, 1904 questionnaires had been returned, for an overall response rate of 33% (Table 1). The response rate was slightly higher for nurses (36%). Among the paramedical specialties the response rate was similar. Seventy-nine percent of the respondents were practicing ICPs in acute care hospitals. Of the ICPs responding, 95% were willing to participate, and 92% had either great or moderate en-
Volume 12 Number April, 1984
2
Needlestick exposures to AIDS
71
APK: NaUonal Regbty of Pamonnol wWhWoa@alExpoalntoAlos Feasibility Questionnaire Today’8 date
Your name (lint)
Are you currently working as an Infecdon Control Practitioner in an acute care hospital 7 Yea - No _ (ff ‘nk’: rhsquwrbnnh Is tmpbd. mw rstumn wnh IK)mfmr myoma.) Name of hospital
Location of hospital:
w
> -m
I .stem -,
zrp-
please record the NUMBER OF AIDS EWOSURES occurring in yuur mtal using the categoric Provided in the table. Count only rigniftcant blood exposures: needlestick or other percutaneous exposure. blood on a fresh (-24 hrs) cut or abrasion, or blood contact with mucous membrane. CATEGORY
OF AIDS-RISK
OF THE “OONOR” OIhw hiih AIDSrisk ‘“z:z IV
MwIe CDC cnDrl8
5mmIh irmnnl durinQ
Hehkm,msic hcmwexusl, hImI of AIDS-risk mother)
f0r NDBchtk 0, Kapwre -8)
Jsn -Juno
1221
July-Dac
1221
Jan.
112
Juna
July-D.0
IS22
Jan 1282 Pmaont
For each future exposure would you bs willing to provide 4 serum specimens Noquestionnaires as outlined in the accompanying letter 7 Yes-
and complete
Is your
little-7
degree
Other comments
of enthusiam
for
this
project
great-7
moderate-7
the
none-?
:
P~~nd~umqwrtbnnWIbyJun~lO.i089~~mAlDS~hoaxlumdinywr~.P~ ueculc-Mdmewd w~anuH~guatlon~rnAPK:Rag*by.Box421MayoBuHding.Univrnity of Mlnnwo~ HoepW. Minmapdir. MN 55455. Thank you.
Fig.
1. Questionnaire mailed to the APIC membership in early June, 1983
thusiasm for the registry project. Enthusiasm for participation was even greater (97%) among ICPs reporting exposures. Willingness to participate and degree of enthusiasm were similar for the various specialties. Since AIDS cases are so highly concentrated in coastal metropolitan areas, response rates were calculated for the high-AIDS incidence areas (Table 2). Standard metropolitan statistical areas (SMSAs) were used to mitigate discrepancies arising when suburban patients use inner city hospitals. Responses were calculated for hospitals with more than 2.50 beds to sim-
plify the computations. Responses accounted for 42% of U.S. hospitals with more than 250 beds. With the possible exception of the Los Angeles-Long Beach SMSA, there appears to be no regional differences in the response rate or willingness to participate. This conclusion holds true whether the responses are considered by hospitals or by the number of beds in those hospitals. Responding APIC members reported that 157 significant blood exposures to blood from patients with AIDS and 43 to patients with lymphadenopathy syndrome had occurred in their
72
Amencan INFECTION
Rhame
Table
1. Analysis
of questionnaire
responses
by specialty
of APIC member Re#ponderr who are prlmtkhlg ICPS
Responses U.S. APIC No.
IlWllbWS
Spscialty
Nurse Physician Labratorian
Other hospital
Journal of CONTROL
%
No.
% a2 77
ICPS wwng to f--Pa@ w
Degee
of
w-m-t w
4362 381 351 42
1568 102 92 8
36 26
1284 79
95 98
92 96
26
68
74
19
5
62
100
100
316
24 97 1904*
7.6 25 33
0 67 1503
0 69 79
97 95
94 92
96
97
personnel
Commercial None specified
382 5834
Total *Excludes tExcludes *Includes
178 who didn’t answer this question 66 who didn’t answer this question. 13 responses from persons not on the APIC mahng
Table 2. Response
from ICPs in short-stay
list
hospitals
with 2250
HospIt&
Geographic
area
% of AIDS casas*
New York SMSA San Francisco-Oakland
SMSA Los Angeles-Long Beach SMSA Miami
SMSA
Newark SMSA Other California Other Other Other
New York New Jersey Florida
Other USA Total
beds, by geographic
Win
with 2260 bdr
ICP response
ICPS wilting
No. of hospitals
No.
%
46 12
66 23
23 10
35 43
7
40
14
4 3
19 20 50 66 31 54
10 a 17 24 12 25 463 606
II 28
-1065 1434
SMSA = Standard metropolitan statistical area ‘As of April 18, 1983, data provided by Tom Stacher,
AIDS Activrty,
Centers
hospitals (Table 3). Ninety-six (61%) of the AIDS exposures occurred in 1983. Nineteen (12%) of the AIDS exposures occurred in hospitals with fewer than 250 beds. As expected, the national distribution of AIDS exposures was similar to the distribution of AIDS cases (data not shown). The change from July through December 1982 to January through June 1983 slightly exceeds the doubling of AIDS cases that has been occurring each 6 months.
The responses to the survey indicate that as of June 15, 1983, at least 200 needlestick or
area hoepWswi#h
ICP reepoMie %
d@Weclr
ICPS wng %
%
No. of beds
22 9
96 90
40591 8235
30 46
a9 90
35
9
64
19559
31
69
53 40 34 36 39 45 43 42
8 7 17 22 11 24 398 527
80 88 100 92 92 96 86 a7
8071 8558 20445 27725 12773 24474 487696 658127
59 43 37 42 40 49 44 43
82 93 100 95 93 97 86 a7
for Disease
No.
Control
other significant in-hospital blood exposures~ had occurred involving blood from patients with AIDS or lymphaidenopathy syndrome. The actual number of such exposures is no doubt higher, since 67% of the survey questionnaires were not returned. Projecting the actual number is speculative; it seems likely that- hospitals with no exposures would be less likely to respond. On the other hand, some hosp$tals with many exposures may have found the compilation too burdensome, particularly in the short reporting time requested by the survey. Some others may not have responded because they thought that any response to the survey was a
Volume
12 Number
2
April, 1984
commitment to participate in the registry, should it be funded. It is encouraging that no AIDS cases have been reported after a prospectively recognized in-hospital AIDS blood exposure. That encouragement must be tempered by recognition that only 19 of the reported exposures to blood from AIDS patients and 20 of the exposures to blood from lymphadenopathy syndrome patients occurred more than a year before the survey. The lower number of early exposures may reflect poor recognition of the problem before July 1982 or poor recall at the time of the survey, but probably also reflects the growing pool of living AIDS patients. How likely is it that AIDS transmission will arise from hospital exposures? In the absence of direct information, the best basis for an answer involves projection from what is known about hospital transmission of other agents that are possible models of AIDS transmission. Attention has been focused on the hepatitis B virus, but in 1982 Tabor’O compiled information about transfusion transmission of 13 viruses, three bacterial species (excluding postcollection contamination), and 11 protozoa1 species. His summary did not include the arboviruses or some new findings of possible relevance in the AIDS context.” Three particularly important transmission paradigms, based on both the epidemiology and pathophysiology of AIDS, are hepatitis B, cytomegalovirus and human T-cell leukemia virus (CMV) , (HTLV). The profiles of persons at risk for AIDS (male homosexuals, intravenous drug users, hemophilia patients, Haitian entrants, transfusion recipients, and infants born to women with AIDS or at increased risk for AIDS) and persons at increased risk of acquiring hepatitis B (male homosexuals, intravenous drug users, hemophilia patients, persons raised in certain underdeveloped countries, transfusion recipients, infants born to women with acute hepatitis B or who are HBsAg positive, hemodialysis patients, and hospital workers) match closely. To the extent that AIDS transmission is similar to hepatitis B transmission, acquisition of the disease by hospital personnel will surely occur. Hospital workers with extensive blood contact, working in hospitals with many hepatitis B
Needlestick
exposures
to AIDS
73
Table 3. Reported
significant blood exposures to blood from patients with AIDS or lymphadenopathy syndrome, by g-month interval Diagnosis of exposing patient 6 mo Interval
January-June, 1981 July-December, 1981 January-June, 1982 July-December, 1982 January-June, 1983’ Total ‘Most
responses
Lymphadenopaihy syndrome
AIDS
were
received
1 4 14 42 96 157 by June
5 9 6 8 15 43 15
carriers, can have up to six times the risk of acquiring hepatitis B as the general population.‘* Transmission rates of hepatitis B after needlestick exposure have been reported in large studies at 27%13 and 12%14 of exposed, susceptible nonprotected hospital workers. From the first recognition of AIDS, CMV has been thought to be an important pathophysiologic paradigm of AIDS.15 CMV induces transient immunosuppression, including a reversed T-lymphocyte helper: suppressor ratio, and is lymphocytotrophic. CMV is also a good transmission paradigm of AIDS. Male homosexuals have an extraordinarily high prevalence and high titers of antibody to CMV.16* I7 This may be related to the greater presence of CMV in semen than in urine or saliva.‘8 CMV is transmitted by blood,‘$ from mother to offspring, and is more prevalent in underdeveloped countries. Less is known about CMV seroprevalence in hemophilia patients and intravenous drug users. Fifteen (65%) of 23 tested, coagulation factor concentrate-treated hemophilia patients reported by deShazo et a1.2Owere CMV seropositive when tested by the relatively insensitive complement fixation (CF) method. However, Kaposi’s sarcoma, a possible marker of CMV infection in AIDS patients, has apparently not occurred in hemophilia patients with AIDS.’ All eight tested intravenous drug abusers with AIDS reported by Small et al?’ had CMV antibodies. To the extent that AIDS transmission is similar to CMV transmission, the short-term implications are favorable for hospital workers. De-
American
74
Rhame
spite the high prevalence of CMV excreters in the hospital, especially on transplant wards and in pediatric areas, and despite a recent secure demonstration of infant-to-infant spread,22 it has been difficult to demonstrate that hospital workers are at an increased risk of acquiring CMV.23 The long-term implications for our species, however, would be ominous. Between 30% and 80% of various adult populations ultimately acquire CMV. It is hard to explain the higher seroprevalence of CMV in underdeveloped countries and in children attending U.S. day care centers24 unless close nonsexual contact results in CMV transmission. At the present time the leading candidate for the AIDS agent is a retrovirusT5 possibly HTLV, an HTLV-related retrovirus, or a novel, unrelated retrovirus isolated at the Pasteur InstituteF6 Relatively little is known about the transmission of HTLV. It is more prevalent in the Caribbean and is endemic in southwestern Japan. 27 Familial clustering is present in Jaoccurs,28 and there pan, 27 vertical transmission is some evidence suggesting transmission by blood transfusion.29 From 5% to 19% of U.S. hemophilia patients have antibodies to HTLV membrane antigens compared to less than 1% for most U.S. populations, including healthy laboratory or hospital workers?* Even workers employed in a laboratory studying HTLV appear not to acquire HTLVP’ Although transmission of HTLV to hospital workers has not been demonstrated, blood precautions have been recommended.32 Moreover, because of the relative ease with which human cord T-cells can be infected by HTLV, avoidance of contact between patients with HTLV infection and pregnant staff has been recommended?* There is a discrepancy among these potential paradigms of AIDS transmission. Hepatitis B and CMV can be transmitted by blood transfusion, but only the former seems to be transmitted by needlestick. The explanation may arise because free hepatitis B virus circulates in the blood while CMV appears to be cell-associated in blood. HTLV will probably be found to be more like CMV in this respect. Much should be gained by prospective study of hospital workers with significant blood exposure to AIDS. If such exposures do not
INFECTION
Journal
of
CONTROL.
transmit AIDS, it is critical to demonstrate the fact. If they do, hospital workers would be a population at increased risk and, at the same time, would have a low probability of already having contracted AIDS at the time of initial study. If so, careful follow-up can provide information about the incubation period of AIDS, the relative infectiousness of patients at various stages of AIDS, and the possible AIDS prodromes. Prospective collection of lymphocytes from exposed hospital workers may resolve the difficulty in interpreting low helper : suppressor ratios in AIDS risk groups. Whether such a study is organized through APIC or the Centers for Disease Contro1,33 a high degree of cooperation by eligible participants is desirable. The tabulations were performed by Susan Reaney and Jill Frost; typing by Jo Kill, Nancy Hovanes. and Carol Sorensen; and secretarial support by Eleanor Jelinek. Centers for Disease Control: Acquired immune de& ciency syndrome (AIDS): Precautions for clinical and laboratory staffs. Morbid Mortal Weekly Rep 31:577580, 1982. Duncanson FP, Joline C, Wormser GP, Chiao JW, Cunningham-Rundles S: Risk of exposure to blood of acquired immune deficiency syndrome patients: A prospective study of hospital personnel. Presented at the Tenth Annual Educational Conference of the Association for Practitioners in Infection Control, San Diego, May 1-5, 1983. Harris C, Small CB, Klein RS, et al: Immunodeficiency in female sexual partners of men with the acquired immunodeficiency syndrome. N Engl J h&d 308: f 18i1184, 1983. Roberts S: Medical detectives hunt clues to AIDS outbreak in New York. New York Times, June 7, 1983. Centers for Disease Control: An evaluation of the acquired immunodeliciency syndrome (AIDS) reported in health-care personnel. Morbid Mortal Weekly Rep 32~358-360, 1983. Centers for Disease Control: A cluster of Kaposi’s sarcoma and Pneumocystis carinii pneumonia among homosexual male residents of Los Angeles and Orange Counties, California. Morbid Mortal Weekly Rep 31: 305-307, 1982. Jaffe H: Epidemiological review of AIDS. Presented at the NIH Research Workshop on the Epidemiology of AIDS, Rockville, Md, Sept 12-13, 1983. American Hospital Association: American Hospital Association guide to the health care field. Chicago, 1982, American Hospital Association. U.S. Department of Commerce Bureau of the Census: Statistics for states and metropolitan areas. A preprint from County and City Data Book 1977, Washington, DC, U.S. Government Printing Office Stock No. 003-024-01487-4.
Volume April,
12 Number
2
1984
10. Tabor E: Infectious complications of blood transfusion. New York, 1982, Academic Press Inc. 11. Mortimer PP, Luban NLC, Kelleher JF, Cohen BJ: Transmission of serum parvovims-like virus by clotting-factor concentrate. Lancet 2:482-484, 1983. 12. Dienstag JL, Ryan DM: Occupational exposure to hepatitis B virus in hospital personnel: Infection or immunization? Am J Epidemiol 115:26-39, 1982. 13. Seeff LB, Wright EC, Zimmerman HJ, et al: Type B hepatitis after needlestick exposure: Prevention with hepatitis B immune globulin. Final report of the Veterans Administration Cooperative Study. Ann Intern Med 88:285-293, 1978. 14. Grady GF, Lee VA, Prince AM, et al: Hepatitis B immune globulin for accidental exposures among medical personnel: Final report of a multicenter controlled trial. J Infect Dis 138:625-638, 1978. 15. Centers for Disease Control: Pneumocystis pneumonia-Los Angeles. Morbid Mortal Weekly Rep 30:250252, 1981. 16. Drew WL, Mintz L, Miner RC, Sands M, Ketterer B: Prevalence of cytomegalovirus infection in homosexual men. J Infect Dis 143:188-192, 1981. 17. Ikram H, Prince AM, Baker LN: Cytomegalovirus (CMV) antibody in male homosexuals: A source for CMV immune globulin. VOX Sang 44:173-177, 1983. 18. Lang DJ, Kummer JF: Cytomegalovirus in semen: Observations in selected populations. J Infect Dis 132: 472473, 1975. 19. Lang DJ, Ebert PA, Rodgers BM, Boggess HP, Rixse RS: Reduction of postperfusion cytomegalovirus-infections following the use of the leukocyte depleted blood. Transfusion 17:391-395, 1977. 20. deShazo RD, Andes WA, Nordberg J, Newton J, Daul C, Bozelka B: An immunologic evaluation of hemophiliac patients and their wives. Relationships to the acquired immunodeficiency syndrome. Ann Intern Med 99:159164, 1983. 21. Small CB, Klein RS, Friedland GH, Moll B, Emeson EE, Spigland I: Community-acquired opportunistic infections and defective cellular immunity in heterosexual drug abusers and homosexual men. Am J Med 74:433-441, 1983.
Needlestick exposures to AIDS
75
22. Spector SA: Transmission of cytomegalovirus among infants in hospital documented by restriction-endonuclease-digestion analyses. Lancet 1:378-381, 1983. 23. Rhame FS: Cytomegalovirus. In Soule BM, editor: The APIC curriculum for infection control practice. Dubuque, 1983, Kendall Hunt Publishing Co, p 433-437. 24. Pass RF, August AM, Dworsky M, Reynolds DW: Cytomegalovirus infection in a day-care center. N Engl J Med 307:477-479, 1982. 25. Maurice J: AIDS investigators identify second retrovirus (medical news). JAMA 25O:lOlO. 1011, 1015, 1021, 1983. 26. Barre-Sinoussi F, Chermann JC, Rey F, et al: Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 220:868-871, 1983. 27. Blattner WA, Takatsuki K, Gallo RC: Human T-cell leukemia-lymphoma virus and adult T-cell leukemia. JAMA 250:1074-1080, 1983. 28. Komuro A, Hayami M, Fujii H, Miyahara S, Hirayama M: Vertical transmission of adult T-cell leukemia virus (letter). Lancet 1:240, 1983. 29. Shimoyama M, Minato K, Tobinai K, et al: AntiATLA (antibody to the adult T-cell leukemia cell-associated antigen)-positive hematologic malignancies in the Kanto district. Jpn J Clin Oncol 12:109-116, 1982. 30. Essex M, McLane MF, Lee TH, et al: Antibodies to human T-cell leukemia virus membrane antigens (HTLV-MA) in hemophiliacs. Science 221: 1061-1064, 1983. 31. Saxinger WC, Blayney DW, Postal M, Blattner WA, Moghissi J, Gallo RC: Risk of infection of laboratory workers with the human T-cell leukemia virus (letter). Lancet 2274-275. 1983. 32. Blayney DW, Blattner WA, Gallo RC: Isolation procedures for patients with leukemia or lymphoma associated with human T-cell leukemia virus (letter). N Engl J Med 308~844, 1983. 33. Centers for Disease Control: Update: Acquired immunodeficiency syndrome (AIDS)-United States. Morbid Mortal Weekly Rep 32:389-391, 1983.
Journal
Volume
Frank S. Rhame, M.D., Minneapolis, Minnesota and the membership of the Association for Practitioners
There have (AIDS) in assess the Practitioners percent of more than
of infection
12 Number
2
Control April
1984
in infection Control
been no documented cases of acquired immunodeficiency syndrome personnel after a prospectively recognized in-hospital blood exposure. To frequency of such exposure the membership of the Association for in Infection Control was surveyed in early June 1983. Thirty-three the membership responded, accounting for 42% of U.S. hospitals with 250 beds. Respondents reported needlestick or other significant blood
exposures to blood from patients with AIDS occurring in 157 instances and to blood from patients with lymphadenopathy syndrome in 43 instances. Nineteen and twenty of the exposures, respectively, occurred before July 1982. There are two bases for believing that AIDS will not pose a substantial risk to hospital workers: the lack of demonstrated in-hospital AIDS transmission to date and the recognition that other viruses besides the hepatitis B virus-viruses that seem to have less potential for in-hospital transmission-are equally plausible models of AIDS transmission. (AM J INFECT CONTROL 12:69-75, 1984.)
Although there is still no published case of clearly established in-hospital transmission of acquired immunodeficiency syndrome (AIDS), there is great concern about potential hazards From the Infection Control Program, University of Minnesota Hospitals and Clinics, and the Infectious Disease Section, Department of Medicine and Department of Laboratory Medicine and Pathology, University of Minnesota. Reprint requests: Frank S. Rhame, M.D., ing, University of Minnesota Hospitals neapolis, MN 55455.
Box 421 Mayo and Clinics,
BuildMin-
to hospital workers from AIDS patients. A valid basis for this concern is the close correspondence between the groups at risk of acquiring AIDS and those at increased risk of acquiring hepatitis B. Hepatitis B is the major hazard associated with hospital work, and AIDS patients tend to be high-acuity patients with great potential for blood exposure. In the hospital, blood has been the only important vector of hepatitis B transmission, and reduction of exposure to blood and blood-containing materials 69
Amerlcan
70
Rhame
has been considered the prime method of preventing transmission of AIDS.’ Thus needlestick exposure is a logical first place to look for evidence of AIDS transmission in the hospital. Direct evidence bearing on needlestick transmission of AIDS is scant. Relative nontransmissibility of AIDS to hospital workers is suggested by one preliminary study.2 Eight personnel had needlestick blood exposure between August and December of 1982. As of May, 1983, careful evaluation, including studies of cell-mediated immunity, showed no sign of AIDS transmission. A needle puncture was reported by one of seven heterosexual women who were regular sexual partners of male AIDS patients,3 but the woman showed no evidence of having acquired AIDS. A dase of AIDS has been reported in a sanitation worker with no risk of exposure to AIDS.4 He worked in the South Bronx, an area with many addicts. No accidental needlestick was recognized, but such exposure may have occurred. Four cases of AIDS in hospital workers with no recognized AIDS risk factors have recently been reported.5 However, only one was suggestive of hospital transmission: the employee worked as a housekeeper in an ambulatory surgery patient area in a Baltimore hospital. He had frequent exposure to blood and had had a needlestick caused by a discarded needle from an unknown source about 15 months before the diagnosis of AIDS. His negative history of other AIDS risk factors was partially confirmed. Information about the other hospital workers was less complete, but one was from the Caribbean and the other two had some suggestion of unconceded exposure to AIDS. None had a recognized exposure to a patient with AIDS. The lack of a single case of AIDS clearly resulting from hospital transmission is encouraging, but proper interpretation of this observation is hampered by two main problems: the long incubation period of AIDS and lack of denominator data describing hospital exposure to AIDS. The incubation period of AIDS in adults is generally considered to be at least a year. In a cluster of AIDS cases occurring in Orange County, California,G three patients with specific sexual contacts with AIDS patients developed symptoms 9, 13, and 22 months, respectively,
INFECTION
Journal
oi
CONTROL
after the contacts. Among 15 adults with AIDS whose only AIDS risk factor was blood transfusion, the mean interval from transfusion to onset of symptoms was 18 months.7 With respect to the lack of denominator data, the study reported herein provides some help. It presents information reported in a survey of the membership of the Association for Practitioners in Infection Control (APIC) conducted in June 1983 on the number of in-hospital needlesticks and other significant blood exposures to blood from AIDS patients and patients with iymphadenopathy syndrome.
As part of an effort to determine the feasibility of a national registry of hospital exposure to AIDS, a survey was mailed on June 3 or 6, 1983, to the 5834 APIC members with addresses in the United States. The mailing contained a cover letter describing a proposal to form the APIC National Registry of Hospital Exposure to AIDS, an explanation of the registry organization, and a questionnaire (Fig. 1). Responses to the third column of the questionnaire table, which requested data on exposures to blood from patients in the AIDS risk category, “other high AIDS risk,” were irregular and are not reported in this article. For making compilations, the following criteria were used: the specialty of the APIC member was that provided on the APIC mailing list; the specialty designations were supplied by the applicant at the time of application to APIC, and some had been updated at the time of membership renewal. Hospital information was taken from the 1972 American Hospital Association Guide? Standard metropolitan statistical area boundaries were taken from City and County Data Book.!” -Ts By July 27, 1983, 1904 questionnaires had been returned, for an overall response rate of 33% (Table 1). The response rate was slightly higher for nurses (36%). Among the paramedical specialties the response rate was similar. Seventy-nine percent of the respondents were practicing ICPs in acute care hospitals. Of the ICPs responding, 95% were willing to participate, and 92% had either great or moderate en-
Volume 12 Number April, 1984
2
Needlestick exposures to AIDS
71
APK: NaUonal Regbty of Pamonnol wWhWoa@alExpoalntoAlos Feasibility Questionnaire Today’8 date
Your name (lint)
Are you currently working as an Infecdon Control Practitioner in an acute care hospital 7 Yea - No _ (ff ‘nk’: rhsquwrbnnh Is tmpbd. mw rstumn wnh IK)mfmr myoma.) Name of hospital
Location of hospital:
w
> -m
I .stem -,
zrp-
please record the NUMBER OF AIDS EWOSURES occurring in yuur mtal using the categoric Provided in the table. Count only rigniftcant blood exposures: needlestick or other percutaneous exposure. blood on a fresh (-24 hrs) cut or abrasion, or blood contact with mucous membrane. CATEGORY
OF AIDS-RISK
OF THE “OONOR” OIhw hiih AIDSrisk ‘“z:z IV
MwIe CDC cnDrl8
5mmIh irmnnl durinQ
Hehkm,msic hcmwexusl, hImI of AIDS-risk mother)
f0r NDBchtk 0, Kapwre -8)
Jsn -Juno
1221
July-Dac
1221
Jan.
112
Juna
July-D.0
IS22
Jan 1282 Pmaont
For each future exposure would you bs willing to provide 4 serum specimens Noquestionnaires as outlined in the accompanying letter 7 Yes-
and complete
Is your
little-7
degree
Other comments
of enthusiam
for
this
project
great-7
moderate-7
the
none-?
:
P~~nd~umqwrtbnnWIbyJun~lO.i089~~mAlDS~hoaxlumdinywr~.P~ ueculc-Mdmewd w~anuH~guatlon~rnAPK:Rag*by.Box421MayoBuHding.Univrnity of Mlnnwo~ HoepW. Minmapdir. MN 55455. Thank you.
Fig.
1. Questionnaire mailed to the APIC membership in early June, 1983
thusiasm for the registry project. Enthusiasm for participation was even greater (97%) among ICPs reporting exposures. Willingness to participate and degree of enthusiasm were similar for the various specialties. Since AIDS cases are so highly concentrated in coastal metropolitan areas, response rates were calculated for the high-AIDS incidence areas (Table 2). Standard metropolitan statistical areas (SMSAs) were used to mitigate discrepancies arising when suburban patients use inner city hospitals. Responses were calculated for hospitals with more than 2.50 beds to sim-
plify the computations. Responses accounted for 42% of U.S. hospitals with more than 250 beds. With the possible exception of the Los Angeles-Long Beach SMSA, there appears to be no regional differences in the response rate or willingness to participate. This conclusion holds true whether the responses are considered by hospitals or by the number of beds in those hospitals. Responding APIC members reported that 157 significant blood exposures to blood from patients with AIDS and 43 to patients with lymphadenopathy syndrome had occurred in their
72
Amencan INFECTION
Rhame
Table
1. Analysis
of questionnaire
responses
by specialty
of APIC member Re#ponderr who are prlmtkhlg ICPS
Responses U.S. APIC No.
IlWllbWS
Spscialty
Nurse Physician Labratorian
Other hospital
Journal of CONTROL
%
No.
% a2 77
ICPS wwng to f--Pa@ w
Degee
of
w-m-t w
4362 381 351 42
1568 102 92 8
36 26
1284 79
95 98
92 96
26
68
74
19
5
62
100
100
316
24 97 1904*
7.6 25 33
0 67 1503
0 69 79
97 95
94 92
96
97
personnel
Commercial None specified
382 5834
Total *Excludes tExcludes *Includes
178 who didn’t answer this question 66 who didn’t answer this question. 13 responses from persons not on the APIC mahng
Table 2. Response
from ICPs in short-stay
list
hospitals
with 2250
HospIt&
Geographic
area
% of AIDS casas*
New York SMSA San Francisco-Oakland
SMSA Los Angeles-Long Beach SMSA Miami
SMSA
Newark SMSA Other California Other Other Other
New York New Jersey Florida
Other USA Total
beds, by geographic
Win
with 2260 bdr
ICP response
ICPS wilting
No. of hospitals
No.
%
46 12
66 23
23 10
35 43
7
40
14
4 3
19 20 50 66 31 54
10 a 17 24 12 25 463 606
II 28
-1065 1434
SMSA = Standard metropolitan statistical area ‘As of April 18, 1983, data provided by Tom Stacher,
AIDS Activrty,
Centers
hospitals (Table 3). Ninety-six (61%) of the AIDS exposures occurred in 1983. Nineteen (12%) of the AIDS exposures occurred in hospitals with fewer than 250 beds. As expected, the national distribution of AIDS exposures was similar to the distribution of AIDS cases (data not shown). The change from July through December 1982 to January through June 1983 slightly exceeds the doubling of AIDS cases that has been occurring each 6 months.
The responses to the survey indicate that as of June 15, 1983, at least 200 needlestick or
area hoepWswi#h
ICP reepoMie %
d@Weclr
ICPS wng %
%
No. of beds
22 9
96 90
40591 8235
30 46
a9 90
35
9
64
19559
31
69
53 40 34 36 39 45 43 42
8 7 17 22 11 24 398 527
80 88 100 92 92 96 86 a7
8071 8558 20445 27725 12773 24474 487696 658127
59 43 37 42 40 49 44 43
82 93 100 95 93 97 86 a7
for Disease
No.
Control
other significant in-hospital blood exposures~ had occurred involving blood from patients with AIDS or lymphaidenopathy syndrome. The actual number of such exposures is no doubt higher, since 67% of the survey questionnaires were not returned. Projecting the actual number is speculative; it seems likely that- hospitals with no exposures would be less likely to respond. On the other hand, some hosp$tals with many exposures may have found the compilation too burdensome, particularly in the short reporting time requested by the survey. Some others may not have responded because they thought that any response to the survey was a
Volume
12 Number
2
April, 1984
commitment to participate in the registry, should it be funded. It is encouraging that no AIDS cases have been reported after a prospectively recognized in-hospital AIDS blood exposure. That encouragement must be tempered by recognition that only 19 of the reported exposures to blood from AIDS patients and 20 of the exposures to blood from lymphadenopathy syndrome patients occurred more than a year before the survey. The lower number of early exposures may reflect poor recognition of the problem before July 1982 or poor recall at the time of the survey, but probably also reflects the growing pool of living AIDS patients. How likely is it that AIDS transmission will arise from hospital exposures? In the absence of direct information, the best basis for an answer involves projection from what is known about hospital transmission of other agents that are possible models of AIDS transmission. Attention has been focused on the hepatitis B virus, but in 1982 Tabor’O compiled information about transfusion transmission of 13 viruses, three bacterial species (excluding postcollection contamination), and 11 protozoa1 species. His summary did not include the arboviruses or some new findings of possible relevance in the AIDS context.” Three particularly important transmission paradigms, based on both the epidemiology and pathophysiology of AIDS, are hepatitis B, cytomegalovirus and human T-cell leukemia virus (CMV) , (HTLV). The profiles of persons at risk for AIDS (male homosexuals, intravenous drug users, hemophilia patients, Haitian entrants, transfusion recipients, and infants born to women with AIDS or at increased risk for AIDS) and persons at increased risk of acquiring hepatitis B (male homosexuals, intravenous drug users, hemophilia patients, persons raised in certain underdeveloped countries, transfusion recipients, infants born to women with acute hepatitis B or who are HBsAg positive, hemodialysis patients, and hospital workers) match closely. To the extent that AIDS transmission is similar to hepatitis B transmission, acquisition of the disease by hospital personnel will surely occur. Hospital workers with extensive blood contact, working in hospitals with many hepatitis B
Needlestick
exposures
to AIDS
73
Table 3. Reported
significant blood exposures to blood from patients with AIDS or lymphadenopathy syndrome, by g-month interval Diagnosis of exposing patient 6 mo Interval
January-June, 1981 July-December, 1981 January-June, 1982 July-December, 1982 January-June, 1983’ Total ‘Most
responses
Lymphadenopaihy syndrome
AIDS
were
received
1 4 14 42 96 157 by June
5 9 6 8 15 43 15
carriers, can have up to six times the risk of acquiring hepatitis B as the general population.‘* Transmission rates of hepatitis B after needlestick exposure have been reported in large studies at 27%13 and 12%14 of exposed, susceptible nonprotected hospital workers. From the first recognition of AIDS, CMV has been thought to be an important pathophysiologic paradigm of AIDS.15 CMV induces transient immunosuppression, including a reversed T-lymphocyte helper: suppressor ratio, and is lymphocytotrophic. CMV is also a good transmission paradigm of AIDS. Male homosexuals have an extraordinarily high prevalence and high titers of antibody to CMV.16* I7 This may be related to the greater presence of CMV in semen than in urine or saliva.‘8 CMV is transmitted by blood,‘$ from mother to offspring, and is more prevalent in underdeveloped countries. Less is known about CMV seroprevalence in hemophilia patients and intravenous drug users. Fifteen (65%) of 23 tested, coagulation factor concentrate-treated hemophilia patients reported by deShazo et a1.2Owere CMV seropositive when tested by the relatively insensitive complement fixation (CF) method. However, Kaposi’s sarcoma, a possible marker of CMV infection in AIDS patients, has apparently not occurred in hemophilia patients with AIDS.’ All eight tested intravenous drug abusers with AIDS reported by Small et al?’ had CMV antibodies. To the extent that AIDS transmission is similar to CMV transmission, the short-term implications are favorable for hospital workers. De-
American
74
Rhame
spite the high prevalence of CMV excreters in the hospital, especially on transplant wards and in pediatric areas, and despite a recent secure demonstration of infant-to-infant spread,22 it has been difficult to demonstrate that hospital workers are at an increased risk of acquiring CMV.23 The long-term implications for our species, however, would be ominous. Between 30% and 80% of various adult populations ultimately acquire CMV. It is hard to explain the higher seroprevalence of CMV in underdeveloped countries and in children attending U.S. day care centers24 unless close nonsexual contact results in CMV transmission. At the present time the leading candidate for the AIDS agent is a retrovirusT5 possibly HTLV, an HTLV-related retrovirus, or a novel, unrelated retrovirus isolated at the Pasteur InstituteF6 Relatively little is known about the transmission of HTLV. It is more prevalent in the Caribbean and is endemic in southwestern Japan. 27 Familial clustering is present in Jaoccurs,28 and there pan, 27 vertical transmission is some evidence suggesting transmission by blood transfusion.29 From 5% to 19% of U.S. hemophilia patients have antibodies to HTLV membrane antigens compared to less than 1% for most U.S. populations, including healthy laboratory or hospital workers?* Even workers employed in a laboratory studying HTLV appear not to acquire HTLVP’ Although transmission of HTLV to hospital workers has not been demonstrated, blood precautions have been recommended.32 Moreover, because of the relative ease with which human cord T-cells can be infected by HTLV, avoidance of contact between patients with HTLV infection and pregnant staff has been recommended?* There is a discrepancy among these potential paradigms of AIDS transmission. Hepatitis B and CMV can be transmitted by blood transfusion, but only the former seems to be transmitted by needlestick. The explanation may arise because free hepatitis B virus circulates in the blood while CMV appears to be cell-associated in blood. HTLV will probably be found to be more like CMV in this respect. Much should be gained by prospective study of hospital workers with significant blood exposure to AIDS. If such exposures do not
INFECTION
Journal
of
CONTROL.
transmit AIDS, it is critical to demonstrate the fact. If they do, hospital workers would be a population at increased risk and, at the same time, would have a low probability of already having contracted AIDS at the time of initial study. If so, careful follow-up can provide information about the incubation period of AIDS, the relative infectiousness of patients at various stages of AIDS, and the possible AIDS prodromes. Prospective collection of lymphocytes from exposed hospital workers may resolve the difficulty in interpreting low helper : suppressor ratios in AIDS risk groups. Whether such a study is organized through APIC or the Centers for Disease Contro1,33 a high degree of cooperation by eligible participants is desirable. The tabulations were performed by Susan Reaney and Jill Frost; typing by Jo Kill, Nancy Hovanes. and Carol Sorensen; and secretarial support by Eleanor Jelinek. Centers for Disease Control: Acquired immune de& ciency syndrome (AIDS): Precautions for clinical and laboratory staffs. Morbid Mortal Weekly Rep 31:577580, 1982. Duncanson FP, Joline C, Wormser GP, Chiao JW, Cunningham-Rundles S: Risk of exposure to blood of acquired immune deficiency syndrome patients: A prospective study of hospital personnel. Presented at the Tenth Annual Educational Conference of the Association for Practitioners in Infection Control, San Diego, May 1-5, 1983. Harris C, Small CB, Klein RS, et al: Immunodeficiency in female sexual partners of men with the acquired immunodeficiency syndrome. N Engl J h&d 308: f 18i1184, 1983. Roberts S: Medical detectives hunt clues to AIDS outbreak in New York. New York Times, June 7, 1983. Centers for Disease Control: An evaluation of the acquired immunodeliciency syndrome (AIDS) reported in health-care personnel. Morbid Mortal Weekly Rep 32~358-360, 1983. Centers for Disease Control: A cluster of Kaposi’s sarcoma and Pneumocystis carinii pneumonia among homosexual male residents of Los Angeles and Orange Counties, California. Morbid Mortal Weekly Rep 31: 305-307, 1982. Jaffe H: Epidemiological review of AIDS. Presented at the NIH Research Workshop on the Epidemiology of AIDS, Rockville, Md, Sept 12-13, 1983. American Hospital Association: American Hospital Association guide to the health care field. Chicago, 1982, American Hospital Association. U.S. Department of Commerce Bureau of the Census: Statistics for states and metropolitan areas. A preprint from County and City Data Book 1977, Washington, DC, U.S. Government Printing Office Stock No. 003-024-01487-4.
Volume April,
12 Number
2
1984
10. Tabor E: Infectious complications of blood transfusion. New York, 1982, Academic Press Inc. 11. Mortimer PP, Luban NLC, Kelleher JF, Cohen BJ: Transmission of serum parvovims-like virus by clotting-factor concentrate. Lancet 2:482-484, 1983. 12. Dienstag JL, Ryan DM: Occupational exposure to hepatitis B virus in hospital personnel: Infection or immunization? Am J Epidemiol 115:26-39, 1982. 13. Seeff LB, Wright EC, Zimmerman HJ, et al: Type B hepatitis after needlestick exposure: Prevention with hepatitis B immune globulin. Final report of the Veterans Administration Cooperative Study. Ann Intern Med 88:285-293, 1978. 14. Grady GF, Lee VA, Prince AM, et al: Hepatitis B immune globulin for accidental exposures among medical personnel: Final report of a multicenter controlled trial. J Infect Dis 138:625-638, 1978. 15. Centers for Disease Control: Pneumocystis pneumonia-Los Angeles. Morbid Mortal Weekly Rep 30:250252, 1981. 16. Drew WL, Mintz L, Miner RC, Sands M, Ketterer B: Prevalence of cytomegalovirus infection in homosexual men. J Infect Dis 143:188-192, 1981. 17. Ikram H, Prince AM, Baker LN: Cytomegalovirus (CMV) antibody in male homosexuals: A source for CMV immune globulin. VOX Sang 44:173-177, 1983. 18. Lang DJ, Kummer JF: Cytomegalovirus in semen: Observations in selected populations. J Infect Dis 132: 472473, 1975. 19. Lang DJ, Ebert PA, Rodgers BM, Boggess HP, Rixse RS: Reduction of postperfusion cytomegalovirus-infections following the use of the leukocyte depleted blood. Transfusion 17:391-395, 1977. 20. deShazo RD, Andes WA, Nordberg J, Newton J, Daul C, Bozelka B: An immunologic evaluation of hemophiliac patients and their wives. Relationships to the acquired immunodeficiency syndrome. Ann Intern Med 99:159164, 1983. 21. Small CB, Klein RS, Friedland GH, Moll B, Emeson EE, Spigland I: Community-acquired opportunistic infections and defective cellular immunity in heterosexual drug abusers and homosexual men. Am J Med 74:433-441, 1983.
Needlestick exposures to AIDS
75
22. Spector SA: Transmission of cytomegalovirus among infants in hospital documented by restriction-endonuclease-digestion analyses. Lancet 1:378-381, 1983. 23. Rhame FS: Cytomegalovirus. In Soule BM, editor: The APIC curriculum for infection control practice. Dubuque, 1983, Kendall Hunt Publishing Co, p 433-437. 24. Pass RF, August AM, Dworsky M, Reynolds DW: Cytomegalovirus infection in a day-care center. N Engl J Med 307:477-479, 1982. 25. Maurice J: AIDS investigators identify second retrovirus (medical news). JAMA 25O:lOlO. 1011, 1015, 1021, 1983. 26. Barre-Sinoussi F, Chermann JC, Rey F, et al: Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 220:868-871, 1983. 27. Blattner WA, Takatsuki K, Gallo RC: Human T-cell leukemia-lymphoma virus and adult T-cell leukemia. JAMA 250:1074-1080, 1983. 28. Komuro A, Hayami M, Fujii H, Miyahara S, Hirayama M: Vertical transmission of adult T-cell leukemia virus (letter). Lancet 1:240, 1983. 29. Shimoyama M, Minato K, Tobinai K, et al: AntiATLA (antibody to the adult T-cell leukemia cell-associated antigen)-positive hematologic malignancies in the Kanto district. Jpn J Clin Oncol 12:109-116, 1982. 30. Essex M, McLane MF, Lee TH, et al: Antibodies to human T-cell leukemia virus membrane antigens (HTLV-MA) in hemophiliacs. Science 221: 1061-1064, 1983. 31. Saxinger WC, Blayney DW, Postal M, Blattner WA, Moghissi J, Gallo RC: Risk of infection of laboratory workers with the human T-cell leukemia virus (letter). Lancet 2274-275. 1983. 32. Blayney DW, Blattner WA, Gallo RC: Isolation procedures for patients with leukemia or lymphoma associated with human T-cell leukemia virus (letter). N Engl J Med 308~844, 1983. 33. Centers for Disease Control: Update: Acquired immunodeficiency syndrome (AIDS)-United States. Morbid Mortal Weekly Rep 32:389-391, 1983.