There has been considerable debate concerning the relative contribution of different modes of transmission to the HIV-1 epidemic in Africa [1–6]. The main mode of transmission, especially in resource-poor settings, is considered to be unprotected heterosexual sex. However, one group of researchers believe that the role of iatrogenic transmission, particularly unsafe medical injections, has been vastly underestimated [1,7]. Accurate estimates of transmission probabilities for contaminated injections are currently lacking, but are needed to assess the risk of infection through this route, and its contribution to the HIV/AIDS pandemic, within resource-poor settings in particular [8,9]. More data are available for single, accidental percutaneous injuries with needles or other sharp objects contaminated with infectious material from HIV patients, usually among health care workers and laboratory workers. This information can help in estimating the risk per contaminated medical injection. A limited number of studies have estimated the risk of HIV transmission per injection drug user (IDU) injection; this information can also help estimate the risk from medical injections, as well as the risk from IDU. The other main iatrogenic risk is unsafe blood transfusions. Therefore in this paper we present the results of a systematic review and meta-analysis on HIV-1 infectivity per single exposure for parenteral modes of transmission (per contaminated medical injection, IDU injection and accidental percutaneous exposure) and blood product transfusion. The results can be used to assess the relative contributions of these modes of transmission.
The systematic review was undertaken according to the QUORUM checklist , revised for review of non-randomized control trial studies.
For parenteral modes of transmission, a search was performed using the criteria for publication titles to contain: HIV AND (infectivity OR infectiousness OR transmission OR probability) AND (parenteral OR occupational OR percutaneous OR needle OR drug use OR intravenous OR health care). Databases used were PubMed MEDLINE at PubMed (National Library of Medicine (NLM) http://www.ncbi.nlm.nih.gov/entrez/query.fcgi), EMBASE (http://gateway.ovid.com/athens) and Science Direct www.sciencedirect.com. Parenteral exposure is defined as subcutaneous, intramuscular or intravenous contact with blood or other body fluid of an HIV-1 infected individual, but not mucocutaneous contact. Percutaneous injuries involving splashes of infected fluid onto open wounds were excluded. For blood transmission, the parenterally-related terms were exchanged for: iatrogenic OR blood OR transfusion OR haemophilia OR hemophilia.
Titles (and abstracts, where available) were scanned for relevance, identifying papers requiring further consideration. Bibliographies of all relevant articles were checked for additional relevant publications. There was no restriction by study design or language of publication.
Exclusion criteria were all abstracts-only pre-1995 and infectivity estimates based on sample sizes of less than 10. Where the most recent report of a prospective study was an abstract, further study information was also taken from the most recent preceding publication, with both abstract and publication cited in the tables. A report by the Public Health Laboratory Service [11,12] was useful to decipher truly independent results and to obtain details on studies for which primary sources were unavailable.
Quantitative data synthesis
To improve comparability of studies, all confidence intervals were recalculated using the Wilson ‘score’ method without continuity correction  as recommended by Newcombe , (except for Hudgens et al. [15,16]). Forest plots were constructed using MS Excel (Microsoft Corp., Redmond, Washington, US).
For accidental percutaneous and blood product exposures, summary estimates and associated 95% confidence intervals (95% CI; based on a fixed effects model) are presented on the forest plots. Two different weighting schemes [sample size (sz) or precision (p)] were used for comparison. Summary estimates used sample sizes for weights (referred to as pooled estimates, psz) or used the inverse variance method  with the precision, defined by the wider bound of the asymmetric Wilson ‘score’ confidence interval [upper bound for needlestick (SiUB) and lower bound for blood (SiLB) exposure], divided by z, used for weights rather than the standard error. This is because needlestick transmission risk is small and blood transmission risk is large, so many studies with 0% (100%) seroconversion for needlestick (blood transfusion) exposure give a standard error of zero and infinite weights.
for needlestick transmission, and
for blood transmission; where ni is the sample size for study, and z = 1.96 denotes the standard normal deviate associated with a two-tailed probability α = 5%.
We tested heterogeneity between studies using the heterogeneity statistic .
Summary of the search
Sixty estimates for parenteral (medical, IDU and accidental injections) and 14 for blood product exposure were found. After evaluation, 30 articles were included for parenteral exposure [15,16,18–45], providing 26 estimates, and 10 articles were included for blood product exposure [46–55], providing 11 estimates. The 30 included studies providing estimates of parenteral transmission consisted of three papers describing IDU risk [15,16,18], 26 papers providing 21 estimates for accidental percutaneous injury [20–45] (where five publications were accompanied by a more recent abstract from which the estimate was taken [27,28,34,35,37–41,56]), and one reporting both percutaneous injury and contaminated injection risks . The 11 estimates for blood product exposure included two from Ward et al. , but only the estimate in which blood products were proved to be contaminated was included in summary estimates. Forest plots of infectivity estimates are presented in Figs 1 and 2. Summaries of the characteristics of included and excluded studies are available on request.
Nineteen of the 25 estimates were derived from prospective studies (one for IDU and 18 for needlestick injury), in which study participants were typically followed up for about 9 to 12 months following one contact/exposure for needlestick or many injection events for IDUs. Studies tended to be prospective in that health care workers or laboratory workers were encouraged to report exposures, after which they were followed up, rather than enrolment in a study before exposure. Two studies of needlestick transmission were retrospective with health care workers sometimes reporting more than one contact/exposure [20,21]. One was of unspecified design , while Gisselquist  used data from a case–control study  to derive an estimate. The remaining two estimates, one for IDU  and one for contaminated injections , were also indirectly derived from mathematical models (one for each type of exposure).
Injection drug users
Infectivity estimates for IDU ranged from 0.63%  to 2.4%  (median = 0.8%) [15,16,18] (Fig. 1). Kaplan and Heimer favoured their lower estimate of 0.67% per exposure , and Hudgens et al.  concluded that their 0.84% estimate was consistent with this. Given the heterogeneity of estimates between studies and subtypes, it was inappropriate to combine results.
The only study available evaluated the estimated infectivity per contaminated injection as 1.9–6.9%, a very wide range, based on data of limited quality from nosocomial outbreaks in Russia, Romania and Libya  and on a large number of assumptions.
Accidental percutaneous injury
Infectivity estimates for accidental percutaneous injury are more reliable than those for IDU and contaminated injections, primarily because the number of exposures is usually one per person, and the infection status of the index case can be determined relatively easily. In the few instances where an individual experienced more than one exposure, these were counted as separate events in our analysis. All studies calculated transmission probabilities as the proportion of exposure events resulting in transmission, except for Gisselquist , who calculated the HIV transmission probability for deep injuries only, to be used as an approximation for contaminated medical injections. The author makes a large number of assumptions, and deep injuries may involve transfer of more infective fluid than contaminated injections. As no confidence intervals were provided or could be derived, this estimate was excluded from summary estimates.
Infectivity estimates for needlestick exposure from all 22 studies ranged from 0.00% [21–30,32–35,42] to 2.38%  (pp = 0.23%; 95% CI = 0.00–0.46%; psz = 0.24%; 95% CI = 0.14–0.40%, n = 21). Transmission of HIV via this route is uncommon and therefore the majority of studies produced estimates of 0.00% ([21–30,32–35,42], n = 13); the average sample size for these studies was 67 injuries, compared with 644 injuries for studies providing non-zero estimates. When including only studies reporting at least one transmission event and excluding Gisselquist , estimates ranged from 0.18% [36,37] to 2.38%  (pp = 0.25%; 95% CI = 0.01–0.49%). Estimates of included studies with pp and psz gave similar results, but we prefer to report pp which produced wider confidence intervals (Table 1). There was no evidence of heterogeneity between studies (df = 20; P = 1).
Analysis of subsets of studies is shown in Table 1. The estimate based on studies with no other reported risk factors for HIV transmission (pp = 0.13%; 95% CI = 0.00–0.54%) was not significantly different from the overall estimate (pp = 0.23%). The infectivity of symptomatic patients (AIDS or AIDS plus AIDS-related complex patients) approximately doubles (psz) or triples (pp) the risk. However, results should be interpreted with caution because of the small sample sizes.
Blood product transfusion
Infectivity estimates for transmission per transfusion with infected blood products are presented in Fig. 2. Eight of the 10 estimates are based on retrospective and two on longitudinal studies following transfusion with infected blood products [52,55]. Infectivity estimates ranged from 27.0 to 100.0%, with summary estimates of pp = 82.2% (95% CI = 79.0–85.4%) and psz = 80.2% (95% CI = 76.7–83.3%). However, the 10 studies displayed considerable heterogeneity (Q = 198.36; χ92: P < 0.0001). The six studies which could confirm that donations were contaminated [49,50,52–55] were more homogeneous (Q = 5.99; χ62: P = 0.42431) and produced higher infectivity estimates (range, 88.3–100.0%; pp = 92.5%; 95% CI = 89.0–96.1%).
Transmission probabilities are difficult to measure for most modes of transmission, partly because of difficulty in confirming the infected source, and also because they depend on other risk factors, such as HIV viral load  and exposure to antiretroviral therapy . The lack of data for both IDU and contaminated medical injection exposure means that transmission probabilities can only be indirectly derived from studies on HIV transmission per accidental percutaneous injury. The risk of infection from percutaneous accidents depends on factors that influence exposure to a greater volume of blood (e.g. visible blood on the device, procedures involving needle placement directly into a vein or artery, etc), viral load (e.g. clinical status of the source patient, post-exposure prophylaxis), which affect the ‘effective’ viral dose received [57,60]. The type of injection equipment affects infectivity because syringes with detachable needles retain – and probably transfer – substantially more blood than integral cannula syringes with a permanently attached needle [61–63]. Laboratory studies have suggested that a greater volume of blood is transferred by deeper injuries and by hollow-bore needles (especially those with larger gauges) .
The reliability of data from studies estimating infectivity per single contaminated medical injection or IDU injection is affected by more factors than studies on accidental percutaneous exposure. The estimates produced by Kaplan and Heimer  and Hudgens et al. [15,16] are compatible with each other and are based on good quality data, considering the difficulty in obtaining information in this area, with thorough analyses. Since HIV-1 CREØ1AE (formerly subtype E) is predominantly contained in South-east Asia and is potentially more infectious than other types, the adjusted estimate for subtype B, 0.63% (95% CI, 0.41–0.92%) is used as the average risk per IDU injection, which is in close accord with Kaplan and Heimer's estimate of 0.67%. Gisselquist's estimate of transmission risk per contaminated injection  is not compatible with these estimates and the methodology has been criticized . The risk from contaminated injections is likely to be less than for IDU because IDU injections are intravenous, whereas medical injections in sub-Saharan Africa are usually intramuscular, blood contamination of needles after use for intramuscular injection is infrequent , and washing of needles substantially reduces risk .
While deliberate injections may involve more transfer of fluid from the device than for accidental punctures, the fluid may contain less infectious material because health care workers will often have cleaned needles between patients. The combined result of these competing effects is unknown and subjective. The true infectivity of contaminated injections is unlikely to be less risky than accidental injuries or to be more risky than for IDU injections. To reflect this uncertainty we propose a transmission probability estimate for contaminated injections ranging between our estimates for these modes: 0.24–0.65%, with a point estimate being the midpoint of 0.45% risk per exposure. Unlike sexual transmission probabilities, these estimates are unlikely to vary substantially by geographical location. Recent estimates of per contact HIV infectivity for sexual transmission in Africa are comparable to this [67,68]. The frequency of sexual exposure is far higher than the frequency of unsafe injections, and so heterosexual sex is the main mode of transmission, with unsafe injections playing a minor role.
Infectivity estimates for infected blood transfusions are much higher than for other modes of HIV-1 transmission due to the far larger viral dose per exposure than for other routes. Relatively few studies aiming to quantify the infectivity of HIV-1 via blood transfusion were found, perhaps because it was clear from the start that this risk was likely to be extremely high, and because this source of infection was identified and eliminated fairly early in the epidemic, at least in industrialised countries. However, the frequency of unsafe blood transfusions compared to sexual contacts is extremely small.
Infectivity estimates per IDU injection range from 0.63%  to 2.4% , median = 0.8%. The only estimate for medical injections was 1.9–6.9% . The estimate per accidental percutaneous injury is pp = 0.23% (95% CI = 0.00–0.46%). We propose a transmission probability estimate per contaminated injection of 0.45% (range, 0.24–0.65%). The transmission probability per contaminated blood product exposure is pp = 92.5% (95% CI 89.0–96.1%). In view of this and the frequency of unsafe injections and blood transfusion compared to unprotected sexual contacts, the role of iatrogenic transmission in resource-poor settings has not been underestimated.
It is important to provide reliable estimates of transmission probabilities for iatrogenic modes of transmission, in order to predict their relative impact on the HIV/AIDS pandemic. Our estimates are based on the most comprehensive review of the literature and best studies, and will be critical when assessing the contribution of these routes of infection to HIV incidence and designing appropriate prevention strategies.
Sponsorship: R.F.B. is grateful to GlaxoSmithKline for financial support. M.A. is a National Researcher of the Fonds de la recherche en santé du Québec (grant no. 8722).
Potential conflict of interest: R.F.B. was supported by an unrestricted educational grant from GlaxoSmithKline.
1. Gisselquist D, Potterat JJ, Brody S, Vachon F. Let it be sexual: how health care transmission of AIDS in Africa was ignored. Int J STD AIDS 2003; 14:148–161.
2. Piot P, Plummer FA, Mhalu FS, Lamboray JL, Chin J, Mann JM. AIDS: an international perspective. Science 1988; 239:573–579.
3. Wyatt HV. Injections and AIDS. Tropical Doctor 1986; 3:97–98.
4. Imperato PJ. The epidemiology of the acquired immunodeficiency syndrome in Africa. NY State J Med 1986; 86:118–121.
5. Brewer DD, Brody S, Drucker E, Gisselquist D, Minkin SF, Potterat JJ, et al
. Mounting anomalies in the epidemiology of HIV in Africa: cry the beloved paradigm. Int J STD AIDS 2003; 14:144–147.
6. Gisselquist D, Potterat JJ. Confound it: latent lessons from the Mwanza trial of STD treatment to reduce HIV transmission. Int J STD AIDS 2003; 14:179–184.
7. Gisselquist D, Potterat JJ. Heterosexual transmission of HIV in Africa: an empiric estimate. Int J STD AIDS 2003; 14:162–173.
8. Chin J, Sato PA, Mann JM. Projections of HIV infections and AIDS cases to the year 2000. WHO Bull 1990; 68:1–11.
9. Hauri AM, Armstrong GL, Hutin YJ. The global burden of disease attributable to contaminated injections given in health care settings. Int J STD AIDS 2004; 15:7–16.
10. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of reporting of meta-analyses. Lancet 1999; 354:1896–1900.
11. PHLS. Occupational transmission of HIV. Summary of published reports
. Public Health Laboratory Service. 1999.
12. PHLS. Occupational transmission of HIV. Summary of published reports
. Public Health Laboratory Service. 2005.
13. Wilson EB. Probable inference, the law of succession, and statistical inference. J Am Stat Assoc 1927; 22:209–212.
14. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 1998; 17:857–872.
15. Hudgens MG, Longini IM Jr, Choopanya K, Vanichseni S, Kitayaporn D, Mastro TD, et al
. Estimating the HIV-1 transmission probability in injecting drug users in Thailand. Appl Stat 2001; 50:1–14.
16. Hudgens MG, Longini IM Jr, Vanichseni S, Hu DJ, Kitayaporn D, Mock PA, et al
. Subtype-specific transmission probabilities for human immunodeficiency virus type 1 among injecting drug users in Bangkok, Thailand. Am J Epidemiol 2002; 155:159–168.
17. Egger M, Davey Smith D, Altman DG, (editors). Systematic reviews in health care: meta-analysis in context
. 2nd edn. London: BMJ Books; 2001. ch. 15.
18. Kaplan EH, Heimer R. A model-based estimate of HIV infectivity via needle sharing. J Acquir Immune Defic Syndr 1992; 5:1116–1118.
19. Gisselquist DP. Estimating HIV-1 transmission efficiency through unsafe medical injections. Int J STD AIDS 2002; 13:152–159.
20. Weiss SH, Saxinger WC, Rechtman C, Grieco MH, Nadler J, Holman S, et al
. HTLV-III infection among health care workers: association with needle-stick injuries. JAMA 1985; 254:2089–2093.
21. Moss A, Osmond D, Bacchetti P, Gerberding J, Levy J, Carlson J, et al
. Risk of seroconversion for acquired immunodeficiency syndrome (AIDS) in San Francisco health workers. J Occup Med 1986; 28:821–824.
22. Kuhls TL, Viker S, Parris NB, Garakian A, Sullivan-Bolyai J, Cherry JD. Occupational risk of HIV, HBV and HSV-2 infections in health care personnel caring for AIDS patients. Am J Public Health 1987; 77:1306–1309.
23. Wormser GP, Joline C, Sivak SL, Arlin ZA. Human immunodeficiency virus infections: considerations for health care workers. Bull NY Acad Med 1988; 64:203–215.
24. Hernandez E, Puyuelo T, Gatell JM, Barrera JM, Pumarola T. [Occupational risk of infection by human immunodeficiency virus]. Article in Spanish. Med Clin (Barc) 1988; 90:767–768.
25. Weiss SH, Goedert JJ, Gartner S, Popovic M, Waters D, Markham P, et al
. Risk of human immunodeficiency virus (HIV-1) infection among laboratory workers. Science 1988; 239:68–71.
26. Strickler AC. Occupational exposure to HIV infection among health-care workers at the Toronto General Hospital. Can Dis Week Rep 1988; 14:141–146.
27. Jorbeck H, Steinkeller E. 41 cases of accidental infection with HIV-positive blood (in Swedish). Lakartidningen 1988; 85:3044–3045.
28. Jorbeck H, Marland M, Steinkeller E. Accidental exposures to HIV-positive blood among health-care workers in 2 Swedish hospitals.Fifth International Conference on AIDS
, Montreal, Quebec, Canada, 1989 [Abstract A517].
29. Rogers PL, Lane HC, Henderson DK, Parrillo J, Masur H. Admission of AIDS patients to a medical intensive care unit: causes and outcome. Crit Care Med 1989; 17:113–117.
30. Puro V, Ranchino M, Profili F. Occupational exposures to blood and risk of HIV transmission in a general hospital. Eur J Epidemiol 1990; 6:67–70.
31. Henderson DK, Fahey BJ, Willy M, Schmitt JM, Carey K, Koziol DE, et al
. Risk for occupational transmission of human immunodeficiency virus type 1 (HIV-1) associated with clinical exposures. Ann Intern Med 1990; 113:740–746.
32. McCormick RD, Meisch MG, Ircink FG, Maki DG. Epidemiology of hospital sharps injuries: a 14-year prospective study in the pre-AIDS and AIDS eras. Am J Med 1991; 91:301S–307S.
33. National surveillance of occupational exposure to the human immunodeficiency virus (HIV).Can Commun Dis Rep
34. Cavalcante NJ, Abreu ES, Fernandes ME, Richtmann R, Piovesana MN, Yamada FT, et al
. Risk of health care professionals acquiring HIV infection in Latin America. AIDS Care 1991; 3:311–316.
35. Cavalcante NJ, Richtmann R, Abreu ES, Ramalho MO, Piovesana MN, Carvalho ES. Changing of notifications of accidents with HIV infective materials, among health care professionals (HCP) at Emilio Ribas Institute, Sao Paolo, Brazil.IX International Conference on AIDS
, Berlin, June 1993 [abstract PO-C18-3030].
36. Ippolito G, Puro V, De Carli G. The risk of occupational human immunodeficiency virus infection in health care workers. Italian Multicenter Study. The Italian Study Group on Occupational Risk of HIV Infection. Arch Intern Med 1993; 153:1451–1458.
37. Ippolito G, De Carli G, Puro V, Petrosillo N, (SIROH). SIROdH. Risk of occupational HIV and HCV infection after occupational exposure.XI International Conference on Aids
, Vancouver, July 1996 [abstract TuC123].
38. Tokars JI, Marcus R, Culver DH. Surveillance of HIV infection and zidovudine use among health care workers after occupational exposure to HIV-infected blood. Ann Intern Med 1993; 118:913–919.
39. Cardo DM, Tokars JI, Marcus R, McKibben PS, et al
. Zidovudine (AZT) use after occupational HIV exposure: toxicity and failures reported to CDC; an update. [Abstract]. Conference on Prevention of Transmission of Bloodborne Pathogens in Surgery and Obstetrics, Atlanta 1994. Infect Control Hosp Epidemiol 1994; 15:342.
40. Communidad de Madrid CdS. Registro de inoculaciones accidentales (VIH) en personal sanitario de la CAM.Vigilancia Epidemiologica del SIDA/VIH
41. de Juanes JR, Lago E, Davila F, et al
. Injuries with exposure to blood in surgeons of a university hospital, over a year period. Abstract. Conference on prevention of transmission of bloodborne pathogens in surgery and obstetrics, Atlanta. Infect Control Hosp Epidemiol 1994; 15:343.
42. Nelsing S, Nielsen TL, Nielsen JO. Occupational exposure to human immunodeficiency virus among health care workers in a Danish hospital. J Infect Dis 1994; 169:478.
43. Gerberding JL. Incidence and prevalence of human immunodeficiency virus, hepatitis B virus, hepatitis C virus, and cytomegalovirus among health care personnel at risk for blood exposure: final report from a longitudinal study. J Infect Dis 1994; 170:1410–1417.
44. Jost J, Iten A, Meylan P, Columbo C, Maziero A. Mise a jour sur les exposition au VIH en milieu medical. Mesures generales, chimioprophylaxie, declaration. Bull Office Fed Sante Pub 1997; 7:5–12.
45. PHLS. Occupational exposures associated with HIV infection in health care workers: update to Sept 97. PHLS AIDS & STD Centre & Scottish Centre for Infection and Environmental Health.Citation from PHLS reports[11,12]
46. Anderson KC, Gorgone BC, Marlink RG, Ferriani R, Essex ME, Benz PM, et al
. Transfusion-acquired human immunodeficiency virus infection among immunocompromised persons. Ann Intern Med 1986; 105:519–527.
47. Kakaiya RM, Cable RG, Keltonic J. Look back: the status of recipients of blood from donors subsequently found to have antibody to HIV. JAMA 1987; 257:1176–1177.
48. Perkins HA, Samson S, Garner J, Echenberg D, Allen JR, Cowan M, et al
. Risk of AIDS for recipients of blood components from donors who subsequently developed AIDS. Blood 1987; 70:1604–1610.
49. Ward JW, Bush TJ, Perkins HA, Lieb LE, Allen JR, Goldfinger D, et al
. The natural history of transfusion-associated infection with human immunodeficiency virus. Factors influencing the rate of progression to disease. N Engl J Med 1989; 321:947–952.
50. Berglund O, Beckman S, Grillner L, Jansson B, Lidbrink P, Karlsson A, et al
. HIV transmission by blood transfusion
in Stockholm 1979–1985 – nearly uniform transmission from infected donors. AIDS 1988; 2:51–54.
51. Gluck D, Kubanek B, Elbert G, Gathof B, Grassmann W, Hesse R, et al
. Risk of HIV infection from former blood donations of donors found to be HIV antibody-positive in blood bank routine testing. ‘Look-back’ study in German Red Cross Blood Banks in the FRG. Infusionstherapie 1990; 17:73–76.
52. Colebunders R, Ryder R, Francis H, Nekwei W, Bahwe Y, Lebughe I, et al
. Seroconversion rate, mortality, and clinical manifestations associated with the receipt of a human immunodeficiency virus-infected blood transfusion
in Kinshasa, Zaire. J Infect Dis 1991; 164:450–456.
53. Donegan E, Lee TH, Operskalski EA, Shaw GM, Kleinman SH, Busch MP, et al
. Transfusion transmission of retroviruses: human T-lymphotrophic virus types I and II compared with human immunodeficiency virus type 1. Transfusion 1994; 34:478–483.
54. Busch MP, Operskalski EA, Mosley JW, Lee TH, Henrard D, Herman S, et al
. Factors influencing human immunodeficiency virus type 1 transmission by blood transfusion
. Transfusion Safety Study Group. J Infect Dis 1996; 174:26–33.
55. Moore A, Herrera G, Nyamongo J, Lackritz E, Granade T, Nahlen B, et al
. Estimated risk of HIV transmission by blood transfusion
in Kenya. Lancet 2001; 358:657–660.
56. Ippolito G, Puro V, De Carli G, Infection. ISGoORoH. Rates of HIV seroconversion by type of exposure: an update of the Italian multicentric study.IX International Conference on AIDS
, Berlin, June 1993 [abstract PO-C18-3021].
57. Cardo DM, Culver DH, Ciesielski CA, Srivastava PU, Marcus R, Abiteboul D, et al
. A case–control study of HIV seroconversion in health care workers after percutaneous exposure. N Engl J Med 1997; 337:1485–1490.
58. Quinn TC, Wawer MJ, Sewankambo N, Serwadda D, Li C, Wabwire-Mangen F, et al
. Viral load and heterosexual transmission of human immunodeficiency virus type 1. Rakai Project Study Group. N Engl J Med 2000; 342:921–929.
59. Fiscus SA, Adimora AA, Schoenbach VJ, Lim W, McKinney R, Rupar D, et al
. Perinatal HIV infection and the effect of zidovudine therapy on transmission in rural and urban counties. JAMA 1996; 275:1483–1488.
60. Gerberding JL. Management of occupational exposures to blood-borne viruses. N Engl J Med 1995; 332:444–451.
61. Zule WA, Ticknor-Stellato KM, Desmond DP, Vogtsberger KN. Evaluation of needle and syringe combinations. J Acquir Immune Defic Syndr Hum Retrovirol 1997; 14:294–295.
62. Grund JP, Friedman SR, Stern LS, Jose B, Neaigus A, Curtis R, et al
. Syringe-mediated drug sharing among injecting drug users: patterns, social context and implications for transmission of blood-borne pathogens. Soc Science Med 1996; 42:691–703.
63. Seamark R, Gaughwin M. Jabs in the dark: injecting equipment found in prisons, and the risks of viral transmission. Aust J Public Health 1994; 18:113–116.
64. Shirazian D, Herzlich BC, Mokhtarian F, Spatoliatore G, Grob D. Needlestick injury: blood, mononuclear cells, and acquired immunodeficiency syndrome. Am J Infect Control 1992; 20:133–137.
65. Schmid GP, Buve A, Mugyenyi P, Garnett GP, Hayes RJ, Williams BG, et al
. Transmission of HIV-1 infection in sub-Saharan Africa and effect of elimination of unsafe injections. Lancet 2004; 363:482–488.
66. Abdala N, Gleghorn AA, Carney JM, Heimer R. Can HIV-1-contaminated syringes be disinfected? Implications for transmission among injection drug users. J Acquir Immune Defic Syndr 2001; 28:487–494.
67. Baeten JM, Richardson BA, Lavreys L, Rakwar JP, Mandaliya K, Bwayo JJ, et al
. Female-to-male infectivity of HIV-1 among circumcised and uncircumcised Kenyan men. J Infect Dis 2005; 191:546–553.
68. Wawer MJ, Gray RH, Sewankambo NK, Serwadda D, Li X, Laeyendecker O, et al
. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005; 191:1403–1409.
69. Kuhls TL, Viker S, Parris NB, Garakian A, Sullivan-Bolyai J, Cherry JD. A prospective cohort study of the occupational risk of AIDS and AIDS-related infections in health care personnel [abstract]. Clin Res 1986; 34:124A.
70. Ramsey KM, Smith EN, Reinarz JA. Prospective evaluation of 44 health care workers exposed to human immunodeficiency virus-1, with one seroconversion [abstract]. Clin Res 1988; 36:119–120.
71. Arva P, Aarseth S, Dobloug JH, Vatn S, Lystad A. Risk of HBV and HIV exposure from injuries by needles and sharp objects among employees in three Norwegian hospitals.International Conference on Blood-borne Infections in the Workplace
, Stockholm, 1989 [abstract no. 5].
72. Pereira LIA, Souza LCS, Souza MA, Silva AB, Oliveira AN. Acidentes profissionais com material biologico de pacientes com sindrome da imunodeficiencia adquirida-acompanhamento clinico-serologico. [abstract 169]. Rev Soc Bras Med Trop 1991; 24:169.
73. Di Vito G, Rizzo F, Poli A, Guida B, Cassola G, Morandi N, et al
. HIV infection and health care workers: results of five years of epidemiologic surveillance in two general hospitals in Genoa, Italy.VII International Conference on AIDS
, Florence, 1991 [abstract WD 4172].
74. Tait DR, Pudifin DJ, Gathiram V, Windsor IM. HIV seroconversions in health care workers, Natal, South Africa.VIII International Conference on AIDS
, Amsterdam, 1992 [abstract PoC 4141].
75. Lot F, Abiteboul D, Bouvet E, Laporte A. Surveillance of occupationally acquired HIV infections in France.IX International Conference on AIDS
, Berlin, June 1993 [abstract PO-C18-3039].
76. Roumeliotou A, Stergiou G, Trichopoulou E, Kotsianopoulou M, Scarpou N, Papaevangelou G. Accidental exposure to HIV of Greek HCW.IX International Conference on AIDS
, Berlin, June 1993 [abstract PO-C18-3032].
77. Inagaki M, Nagao T, Kawasaki N, Shirahata S. Incidence of HIV needlestick accident in health care workers and family members of haemophiliacs on home infusion therapy.X International Conference on AIDS
, Yokohama, Japan, 1994 [abstract PB0967].
78. Castillo TN, Fajardo VR. Prospective study of HCW of a medical center with accidental exposure to patients infected with HIV-1.X International Conference on AIDS
, Yokohama, Japan, 1994 [abstract PC0421].
79. Ippolito G, Puro V, De Carli G, Infection. ISGoORoH. This risk of occupational HIV infection in health care workers: Italian multicentre study update as of September 93. [abstract]. Conference on Precention of Transmission of Bloodborne Pathogens in Surgery and Obstetrics, Atlanta 1994.Infect Control Hosp Epidemiol
80. Ippolito G, Puro V, De Carli G, Infection. ISGoORoH. The risk of occupational HIV infection in health care workers: Italian multicentre study update as of December 1993.III International Conference of the Hospital Infection Society
, London, September 1994 [abstract F6/2].
81. Esteban JI, Shih JW, Tai CC, Bodner AJ, Kay JW, Alter HJ. Importance of western blot analysis in predicting infectivity of anti-HTLV-III/LAV positive blood. Lancet 1985; 2:1083–1086.
82. Menitove JE. Status of recipients of blood from donors subsequently found to have antibody to HIV. N Engl J Med 1986; 315:1095–1096.
83. Kleinman S, Secord K. Risk of human immunodeficiency virus (HIV) transmission by anti-HIV negative blood: estimates using the lookback methodology. Transfusion 1988; 28:499–501.