Post-exposure prophylaxis for HIV infection

Review For reprint orders, please contact [email protected]

Post-exposure prophylaxis for HIV infection Expert Rev. Anti Infect. Ther. 9(4), 431–442 (2011)

David Rey Le Trait d’Union, Centre de Soins de l’Infection par le VIH, NHC, Hôpitaux Universitaires, 1 Place de l’Hôpital, 67091 Strasbourg, France Tel.: +33 036 955 0501 Fax: +33 036 955 1714 [email protected]

Post-exposure prophylaxis (PEP) with antiretrovirals is now widely used worldwide after either occupational contact (with blood or another blood-containing fluid), or nonoccupational contact (mainly sexual or injection-drug use). It is assumed to reduce the risk of HIV transmission by at least 80%, although its efficacy has not been completely proven. Several countries have issued guidelines to help the clinician in their decision to offer PEP and to improve its cost–effectiveness. This article reviews the different antiretroviral combinations used, their safety profile, the recommendations and indications of PEP. The initial evaluation of an exposed individual is presented, as well as the follow-up during and after treatment. Keywords : adherence • adverse events • antiretrovirals • cost–effectiveness • follow-up • HIV testing • lopinavir • post-exposure • prophylaxis • raltegravir • transmission risk

With approximately 2.7 million new cases worldwide annually [101] , including 55,000 in North America [1,101] and 30,000 in Western and Central Europe in 2008 [101] , the HIV infection epidemic is still running and any strategy which has the capacity to prevent new infections has to be developed and optimized. The use of oral antiretrovirals (ARVs) to decrease HIV transmission after potential exposure (post-exposure prophylaxis [PEP]), dates back to the early 1990s, but was not the standard of care, owing to the limited availability of ARV drugs (very few besides zidovudine [ZDV]). A randomized, placebo-controlled trial of ZDV was initiated at that time, but enrolment was not completed [2] . The next and critical step, which resulted in routine prescription of post-exposure ARV therapy, occurred in 1997, when a retrospective study showed, in exposed healthcare workers, an 80% reduction of transmission risk with ZDV, compared with placebo [3] . In the mean time, larger availability of ARV also offered the opportunity to optimize these strategies. Post-exposure treatment has been primarily used after occupational contacts and later extended to all possible exposures to HIV (through sexual contact or after injection-drug use), from a source known to be HIV positive, or also of unknown status. Despite possible nonsevere adverse events and the knowledge that this strategy does not provide full protection, PEP is increasingly widely used and accepted, www.expert-reviews.com

10.1586/ERI.11.20

or requested. Potential exposure to HIV is also a stressful situation for the exposed individual (whatever the mechanism), reinforcing the need for a trained hospital unit for PEP, as an emergency unit taking care of these patients when the HIV specialist is out of the office and, ideally, a telephone referent available 24 h a day. Rationale for PEP possible efficacy

As a randomized, placebo-controlled clinical trial to demonstrate (or not) the efficacy of PEP would now be unethical (mainly because of the suggested efficacy of ZDV in the retrospective study presented below) and nonfeasible for logistical reasons (controlled trials would be too expensive and would require too many subjects, given the low pre-exposure risk of transmission), the use of PEP in the clinical setting is based on: • The results from animal transmission models; • The efficacy of ARV treatment in perinatal clinical trials; • A case–control study on occupational exposure to HIV and observational studies. Animal studies

Animal studies evaluating the potential benefit of PEP have shown conflicting results. Treatment with tenofovir (TDF) prevented acute SIV infection in the macaque model after intravenous inoculation, and both delay of treatment initiation and duration of therapy were crucial factors for efficacy [4] . TDF also has the ability

© 2011 Expert Reviews Ltd

ISSN 1478-7210

431

Review

Rey

to prevent infection after vaginal inoculation in the pig-tailed macaque model, with early intervention being again essential [5] . These data are consistent with the biology of vaginal SIV transmission: SIV enters the vaginal mucosa within 60 min of intravaginal exposure and SIV-infected cells are detected in draining lymph nodes within 18 h [6] . On the other hand, all macaques intravenously inoculated with SIV/HIV chimera became infected despite post-exposure treatment combining ZDV, lamivudine (3TC) and indinavir for 28 days, even when administered as early as 4 h postinoculation [7] . However, plasma viral load was reduced in infected macaques despite receiving HAART, in comparison to control animals [7] . In addition to early treatment initiation, strict compliance to the duration of therapy is a critical factor for the success of prophylaxis [8] . Pre-exposure prophylaxis (PrEP) treatment is now under investigation and has been shown to protect macaques after rectal challenge when Truvada (fixed-dose combination of TDF and emtricitabine [FTC]) was orally administered before and after viral exposure, and a shorter delay between viral challenge and post-exposure drug administration was associated with a higher efficacy [9] . Finally, new ARV drugs such as maraviroc may be interesting as they bind to the CCR5 co-receptor, thus impeding HIV-1 entry into cells. Used as a vaginal microbicide, it showed some protection in a rhesus macaque model, the delay between maraviroc application and virus challenge being critical again [10] . Postnatal prophylaxis

Antiretroviral drugs reduce maternal–infant transmission of HIV, and this has been shown for the first time in 1994. Although decrease in maternal viral load is a major factor in the effectiveness of the treatment, neonate prophylaxis gives an additional effect. A study conducted in Malawi among women without ARV intrapartum treatment compared postnatal single-dose nevirapine (NVP), with and without ZDV for 1 week [11] . Transmission rate was 7.7% among children treated with ZDV + NVP and 12.1% among those who received NVP alone, which is lower to historical rates before the HAART era. A partial protective effect also appears in a South African study comparing single-dose NVP and 6 weeks of ZDV. Both started within 24 h of delivery, in 1051 infants whose mothers had no prior ARV therapy: at 12 weeks, 7.9% infections occurred in the NVP arm and 13.1% in the ZDV arm [12] . HIV transmission rate was 26.6% in nonprophylaxed women in the renowned pediatric ACTG 076 study, when this rate was 9.3% if ZDV was started within the first 48 h of life, increasing to 18.4% if treatment was begun on day 3 of life or later [13] . Case–control & observational studies

Post-exposure prophylaxis efficacy demonstration in the setting of occupational exposure is limited to a case–control study, which estimated that ZDV prophylaxis taken by healthcare workers after exposure reduced the odds of HIV infection by approximately 81% [3] . However, this study suffers several limitations, primarily its retrospective nature and also the small number of case patients and lack of a standardized treatment protocol. 432

Some observational studies have suggested that PEP might reduce the risk of HIV infection after sexual exposure (few of them being published), but they were not randomized and sample sizes were too small to allow statistical conclusions [14] . On the other hand, failure of PEP has also been documented after both occupational [15,16] and nonoccupational exposure [17–19] . Finally, a systematic review recently concluded that it was not possible to draw conclusions on the effectiveness of nonoccupational PEP because of limited available evidence [20] . As PEP prophylaxis has some limitations, other interventions are under investigation, such as PrEP. A TDF vaginal gel reduced HIV transmission rates by 39% among women in South Africa [21] and oral TDF/emtricitabne administered to HIV-negative men or transgender women, who have sex with men, was also shown in a recent multicenter, placebo-controlled trial, to decrease the incidence of HIV by 44% [22] . Resistance testing in infected individuals failed to detect any resistance mutation for TDF or FTC. However, the protective effect was lower than initially hypothesized and measured drug exposure was lower than reported pill use (possibly because of side effects reported in the first month of treatment) [22] . Potential risks of PEP use

As PEP use is, by definition, a prophylactic intervention, it is critical that it is a less normal option. Potential risks could be the increase in high-risk sexual behavior with nonoccupational PEP, the occurrence of possibly severe and eventually life-threatening adverse events due to ARV drugs, as well as the selection of resistant virus, in all kinds of PEP. The use of PEP is only rational if the risks are lower than the benefits. Sexual behavior

It could be argued that availability of nonoccupational PEP would encourage behavioral disinhibition. However, this has not been demonstrated and several studies found a lack of disinhibiton in PEP users. Among 104 heterosexual HIV-serodiscordant couples, over one-quarter of all respondents had heard about PEP and awareness of PEP was not associated with unprotected sex in seronegative or seropositive partners [23] . However, seronegative individuals were more likely to report HIV risk taking in the light of new HIV treatments. An observational study in Amsterdam, The Netherlands, showed a very modest increase in PEP requests in 5 years (2000–2004) [24] , mostly from men who have sex with men (MSM) (75%); while there was no behavioral evaluation, this increase is reassuring. Of all people requesting PEP treatment in 5 years (245), only four started two PEP courses and two started three courses. Another study assessed perceptions and use of nonoccupational PEP over 4 years in six cities across the USA and included 4295 MSM [25] . At baseline 2037 participants had heard of PEP; overall, 1.9% of MSM reported use of nonoccupational PEP prior to enrolment and 6.3% at least once during the trial. Use of PEP was rather low (315 people on at least one occasion): 81 (1.9%) reported PEP use before enrolment and 256 people (6.3%) reported PEP Expert Rev. Anti Infect. Ther. 9(4), (2011)

Post-exposure prophylaxis for HIV infection

use during study participation, 44 of whom reported PEP use at more than one visit. PEP use occurred more frequently in men with high-risk sexual behaviors. For participants who had used PEP prior to the study, the majority did not think the availability of PEP after high-risk exposures would increase the likelihood of unsafe sex. In those who had never used PEP, the majority view was opposite and among those using PEP for the first time during study follow-up, the perception that nonoccupational PEP would increase unsafe sex, changed from the agreement to disagreement range following first PEP use. The authors conclude that PEP was not perceived as an alternative to safe sex, nor that the experience of accessing and taking PEP increased risky sexual behavior. Annual questionnaires about nonoccupational PEP use were collected in 1427 MSM included in Sydney, Australia, in a community cohort from 2001 to 2007 [26] . Awareness of the availability of PEP was high (78.5% of participants at baseline and 97.4% after 5 years). If PEP use increased over time, it was not associated with changes in sexual behavior. Therefore, nonoccupational PEP is mostly used by subjects of higher risk sexual behavior, but its availability does not appear to encourage such conduct. However, the HIV epidemic remains rather high in developed countries, especially in MSM, which suggests that, despite apparent information [26] , these individuals do not request PEP, whatever the reason is, justifying the evaluation of PrEP. Adverse events due to PEP

Clinical experience has shown for several years that clinical side effects are more common during PEP treatment, when compared with HIV-infected patients under HAART. Therefore, it clearly appears that the psychological context of preventive therapy and fears of contamination enhance the frequency of clinical complaints. This is supported by a study comparing adverse events to ZDV/3TC/indinavir, which were observed in only 11% of HIV-infected subjects and 70% of uninfected people receiving PEP [27] . A combination of three ARV drugs is usually recommended, although it has not been shown in the PEP setting to be superior to a two-drug regimen [28] . Protease inhibitor (PI)-containing treatments have been more extensively evaluated and successively studied in parallel with their commercial availability. Fewer adverse effects were reported with atazanavir (ATV) associated to two nucleoside analogues (NAs), compared with nelfinavir and the same two NAs [29] , or with lopinavir/ritonavir (LPV/r) [30] also compared with nelfinavir. LPV/r seems to have a good safety profile, in combination with ZDV/3TC [31] or with TDF/ emtricitabine [32] . This combination appears to be the most widely used in 2010. A higher risk of hepatotoxicity has been observed with fos-amprenavir/ritonavir [33] , which is therefore not recommended. ATV, with or without ritonavir, in combination with two NA, has roughly the same tolerance as LPV/r [34,35] , but due to frequency of jaundice and ATV cost, this combination is not recommended. A regimen consisting of three NAs (ZDV/3TC/ TDF) might be interesting, but has the same tolerability as LPV/r or ATV [35] . www.expert-reviews.com

Review

It must also be remembered that severe illness can occur with LPV/r, as reported a few years ago in Switzerland [36] , with a severe multiorgan failure starting the day after ZDV/3TC and LPV/r initiation in a healthcare worker, who recovered slowly and was unable to resume her work 6 months later. NVP is no longer recommended in the setting of PEP because of hepatic and cutaneous toxicity [37] . However, a recent review showed that rates of hepatotoxicity were lower when NVP was administered for less than 2 weeks [38] , including two studies with a 4-day short course of NVP (200 mg per day) in combination with two NAs [39,40] . Although not recommended in the guidelines, NVP is more interesting than PIs owing to its pretranscriptional mode of action and because it is much more cost efficient than other ARV classes. More recently licensed drugs are now being evaluated and could be promising. Raltegravir (RAL), an integrase inhibitor, well tolerated in HIV-infected subjects, induced few, if any, adverse reactions in three recent case reports [41,42] . In a clinical study including 39 individuals seeking nonoccupational PEP, RAL with TDF/FTC induced symptoms less frequently compared with historical controls who received ZDV/3TC plus a PI [43] . A nonrandomized clinical trial evaluating tolerance of TDF/FTC + RAL in occupational or nonoccupational PEP is planned in France (and started in November 2010). Maraviroc (a CCR5 coreceptor inhibitor) has been used in a single healthcare worker after percutaneous injury to an HIV-infected patient with late-stage disease and mutidrug-resistant virus, in combination with TDF, 3TC, fos-amprenavir/ritonavir and was also well tolerated [44] . Table 1 summarizes clinical trials assessing tolerance of different ARV combinations in the PEP setting. Resistance to ARV induced by PEP

Whether PEP could induce ARV resistance in the exposed individual without avoiding HIV contamination has not been demonstrated. When an individual becomes infected with a resistant virus despite taking PEP [16] , it is more likely a resistant viral strain from the index case that is transmitted, indicating failure of PEP. In one subject infected with HIV-1 after sexual exposure and failure of TDF/FTC prophylaxis [17] , selection of drug resistance did not occur, and an attenuated clinical course and a low seroconversion kinetic were observed, which reduced the probability of subsequent forward transmission. The authors concluded that their results support continued investigations of the use of ARV as a means to reduce HIV-1 transmission. Cost–effectiveness of PEP

Cost–effectiveness of interventions providing PEP after sexual or injection-drug use exposures were evaluated in US programs in San Francisco, CA, USA [45] and 96 metropolitan areas [46] . HIV PEP was cost effective, with cost–utility ratios of US$14,449 and US$12,567 per quality-adjusted life-year (QALY) saved, respectively. In both studies, the majority of HIV infections prevented by PEP were among persons with high-risk behavior (e.g., receptive anal intercourse). By contrast, the French PEP program [47] is only moderately cost effective (overall ratio of €88,692 per QALY 433

Review

Rey

Table 1. Safety of different post-exposure prophylaxis combinations. Antiretroviral combination

Type of exposure Subjects (n)

Subjects experiencing adverse events

Treatment interruptions for side effects (%)

ZDV/3TC + LPV/r ZDV/3TC + ATV 400 mg q.d.

Sexual: 91%

56% 52%

7

[34]

ZDV/3TC + NFV ZDV/3TC + ATV 400 mg q.d.

Sexual (men who 261 have sex with men)

92% (diarrhea: 73%) 82% (diarrhea: 11%)

11 8

[29]

ZDV/3TC + NFV

Nonoccupational

548

58%

5

[30]

108

45%

171

45–50%

18

[35]

152

43–50%

21

ZDV/3TC + LPV/r ZDV/3TC/TDF

All

TDF/3TC + ATV/r

131 124

Ref.

TDF/FTC/RAL

Nonoccupational

39

2–50

[43]

TDF/FTC + LPV/r

All

249

Tolerability Bad: 7% Moderate: 35% Good: 58%

11

[32]

ZDV/3TC + FPV/r

All

26

Rash: three

11–50

[33]

[31]

Grade 4 liver toxicities: two ZDV/3TC + LPV/r

All

135

59% (98 patients who needed 28 days of treatment)

20

ZDV/3TC + NVP

All

120

56–70%

10

[39]

Not indicated

[40]

ZDV/3TC + NVP

All

131

Nausea: 26% Fatigue: 25% Headache: 11%

3TC: Lamivudine; ATV: Atazanavir; LPV: Lopinavir; NFV: Nelfinavir; NVP: Nevirapine; q.d.: Once a day; r: Ritonavir; RAL: Raltegravir; TDF: Tenofovir; ZDV: Zidovudine.

saved). However, there were differences according to the type of exposure. PEP was cost saving in men and women after receptive anal intercourse with an HIV-infected individual, in intravenous drug users having shared needles with an HIV-infected individual, for healthcare workers after percutaneous exposure to material from an HIV-infected patient and for MSM who had receptive anal intercourse with a partner with unknown HIV status. In Australia, nonoccupational PEP was also not cost effective, except after receptive unprotected anal intercourse exposure to an HIV-positive source [48] . Therefore, in developed countries, cost–effectiveness of PEP could be largely improved by targeting high-risk exposure. A systematic review on nonoccupational PEP confirms that it might be cost effective, especially in certain population subgroups [20] . Beyond the cost–effectiveness question, it clearly appears that PEP is increasingly widely used, or requested and often too easily prescribed, which means one guideline not being respected; some recommendations may have resulted in increasing use of PEP for low risk exposures [49] . However, PEP is probably not considered in some high-risk exposures (mostly nonoccupational), usually owing to lack of information). Initial evaluation of a potential exposure

This evaluation requires three steps/questions without losing any time during the process: 434

• Is the exposed individual non-HIV-infected? • Is the exposure at risk of HIV transmission and what is the level of risk? • Is the source HIV-positive? If yes, what is the history of HIV infection? HIV status of the exposed person

Initial HIV testing (rapid test or standard ELISA) should be performed in all exposed persons. When a healthcare worker experiences an occupational exposure, a baseline negative HIV test is mandatory to document a (hopefully rare) professional infection. After nonoccupational exposure, the decision to offer PEP is based on the absence of HIV infection of the exposed individual. Initial HIV testing in this exposed subject should be combined with other blood- or sexually transmitted pathogens: hepatitis B (HBs antibodies, completed with HBs antigen and HBc antibodies if necessary) and hepatitis C (ELISA), as well as syphilis (rapid plasma reagin test or Venereal Disease Research Laboratory + Treponema pallidum hemagglutination assay or ELISA) after sexual exposure. In the absence of vaccination for HBV in the exposed individual, and in case of possible hepatitis B in the source subject, knowledge of anti-HBs result is critical (and therefore should be obtained as quickly as possible), as passive immunization is possible and should not be delayed. Expert Rev. Anti Infect. Ther. 9(4), (2011)


If results of the exposed individual are positive for HIV, PEP will be interrupted. Timing & characteristics of the exposure

When PEP is offered, we have seen previously (animal models) that a critical point for efficacy is the time elapsed between exposure and first administration of ARV drugs. PEP should be initiated within 72 h of exposure [50] , or no later than 48 h after exposure [102] . Moreover, the faster PEP is started, the more effective it will be (ideally within 4 h after exposure). HIV cellular targets are activated CD4 + T cells, dendritic cells or macrophages, therefore needing a percutaneous or mucosal (i.e., vagina, ectocervix, foreskin, rectum or possibly buccal mucosa) contact of an HIV-containing fluid, for possible contamination [51] . At-risk exposure for healthcare workers is defined as a percutaneous injury (needlestick or cut with a sharp object) or contact of mucous membrane or nonintact skin with blood, tissue or other body fluids that are potentially infectious [51] . Semen and vaginal secretions have not been implicated in occupational HIV transmission. The following fluids are also considered potentially infectious: cerebrospinal fluid, synovial fluid, pleural fluid, peritoneal fluid, pericardial fluid and amniotic fluid but the risk for transmission of HIV infection from these fluids is unknown. Feces, nasal secretions, saliva, sputum, sweat, tears, urine and vomitus are not considered potentially infectious unless they are visibly bloody. Mean HIV transmission risk after percutaneous exposure is only 0.3% and after mucous membrane exposure is approximately 0.09%. It increases with exposure to a larger amount of blood from the source person, indicated by visible blood on the device, a needle directly placed in the vein or artery and a deep injury (Tables 2 & 3) . On the other hand, risk is roughly halved when the device crosses a glove, or any other material (i.e., cloth, storage box or trash bag). The estimated per-act transmission risk from sexual exposure (unprotected or after condom breakage) with a partner known to be HIV infected is relatively low, but differs accordingly to sexual practice [52] : if receptive anal intercourse has the highest risk, insertive anal intercourse, penile–vaginal exposures and oral sex represent substantially less per-act risk (Table 2) . A recent study conducted in the HAART era in MSM, showed that despite the fact that a high proportion of HIV-infected men are on ARV treatment and have undetectable viral load, the per-contact probability of HIV transmission due to unprotected anal intercourse is similar to estimates reported from developed country settings in the pre-HAART era [53] . The risk of transmission after receptive anal intercourse decreases (by a ratio of two) if withdrawal occurs before ejaculation [54] . STIs can facilitate HIV-1 transmission by increasing both infectiousness and HIV-1 susceptibility. Indeed, genital ulcer disease results in breached epithelium, monocyte recruitment, increased proinflammatory cytokine and increased HIV-1 genital shedding [53] . Lack of circumcision is also associated with a higher risk of sexual transmission, as three randomized trials have now shown that per-contact transmission risk was reduced by 50–60% in circumcised heterosexual men in an African setting [55–57] . www.expert-reviews.com

Review

Table 2. HIV risk transmission according to exposure type. Type of exposure†

Risk per 1000 exposures to an infected source

Needle-sharing injection-drug use

6.7

Receptive anal intercourse

5

Percutaneous needle exposure

3

Receptive vaginal intercourse

1

Mucous membrane exposure

0.9

Insertive anal intercourse

0.65

Insertive vaginal intercourse

0.5

Receptive oral intercourse

0.1

Insertive oral intercourse

0.05

Estimated per-act risk for acquisition of HIV. Data from [3,52,54,55]. †

On an individual basis and after vaginal intercourse, PEP could be avoided (or stopped after a few days if immediately initiated) when the treated HIV-infected indivdual has a confirmed undetectable plasma HIV viral load for several months, a good adherence to ARV and both partners are free of any other STI. Such a decision could be facilitated if the treated individual is taking ARV having a good diffusion in genital secretions. Among injection drug users, needle sharing has the highest risk of HIV transmission, which is estimated to be 0.63–2.4% per needle-sharing contact [58,59] with a lower transmission (not quantified) risk from sharing other contaminated drug-use equipment. An important cause of HIV infection in people who inject drugs is attributable to sexual transmission [59] . Other nonoccupational exposures [53] include percutaneous injuries from needles discarded in public settings (parks, buses, etc.).�� These �� injuries usually involve small-bore needles that contain only limited amounts of blood and the viability of any virus present is limited. Survival of HIV-1 in syringes is associated with the amount of blood and the storage temperature (at 4°C, 50% of syringes still retained viable HIV after 42 days of storage but, above room temperature, less than 1% of syringes contained viable viruses) [60] . Although HIV transmission is occasionally Table 3. Significant factors for HIV transmission after percutaneous exposure. Characteristics of exposure†

Adjusted odds ratio (95% CI)

Deep injury

15 (6–41)

Visible blood on device

6.2 (2.2–21)

Needle in vein or artery

4.3 (1.7–12)

Terminal illness in source patient

5.6 (2–16)

Post-exposure prophylaxis with zidovudine

0.19 (0.06–0.52)

Risk factors for HIV transmission after percutaneous exposure to HIV-infected blood. Data from [3]. †

435

Review

Rey

reported [61] , the risk after cutaneous blood contact is very low and PEP is usually not recommended (and therefore discussed on a case-by-case basis). Human bites could theoretically be another mode of transmission, but it has been very rarely reported [62] and not reported recently. As saliva (without blood) contains very low titers of virus, being bitten by an HIV-infected person constitutes a negligible exposure risk. Biting an HIV-infected person exposes the oral mucous membranes to infected blood and could put the individual at (low) risk. HIV status of the source person

When the source person is known to be HIV infected, it is critical to evaluate their clinical status, most recent HIV RNA and lymphocyte CD4 + T-cell count results, past and ongoing ARV treatment, and possible HIV resistance to ARV. PEP will then be decided according to exposure characteristics and, if needed, the prophylactic regimen will be adapted to the sources treatment. When the source subject is known to be HIV negative, PEP is usually not indicated, but remains possible when the source acknowledged recent high-risk behavior (therefore a potential recent HIV contamination, with a high viral load). When the HIV status of the source is unknown, it should be determined whether the source is available for HIV testing. It can be usually resolved quickly in occupational settings (unless exposure occurred with some discarded material from an unknown origin) and the result can be obtained shortly with rapid tests. However, after nonoccupational exposure, the source patient is rarely available. PEP decision then is built on exposure risk and the likelihood that the source is known to be from a high-risk group (i.e., MSM, injection-drug users, commercial sex workers, individuals from a country where HIV prevalence is 1% or greater and individuals who have unprotected sexual intercourse with multiple partners). If HIV testing in the source patient must be delayed, it is recommended to start PEP, pending the result of the source. Recommendations

When evaluation of exposure is completed, a recommendation for treatment, or not, has to be formulated to the exposed person (Table 4) .

to a two-drug regimen in a PEP setting, this superiority (based on the higher viral suppression in treating HIV-infected persons and a long clinical experience) is highly probable. Two-drug regimens are rarely recommended by US guidelines after percutaneous injury and are mostly recommended in cases of mucous membrane exposure [51] , and it may also be considered after nonoccupational PEP [53] . European guidelines are simpler and, in the authors opinion, more pragmatic, always recommending, if PEP is offered, a three-drug combination [102] . Combination of two NAs represents the backbone of PEP (similar to treatment in HIV-infected patients), mostly ZDV + 3TC (as a fixed dose combination) or TDF + FTC (also as a fixeddose combination). Although frequently associated with adverse events, they have been the most extensively studied in the PEP setting and other drugs such as stavudine or didanosine should be avoided as they have higher toxicity [64] . The third drug could be a non-nucleoside reverse transcriptase inhibitor or a PI. Among the former, EFV has been recommended in US guidelines (but should not been used in pregnant women), and is likely only rarely prescribed in clinical routine (owing to neuropsychological side effects). NVP is not recommended, despite being effective on a short-course basis and also less expensive. Therefore, PIs are the more common third agents used in PEP, and LPV is the most widely used. European guidelines recommend the combination of TDF/FTC (alternative: ZDV/3TC) and LPV/r while boosted ATV or darunavir, are other options in US recommendations [65] , in addition to LPV/r and in combination with two NAs (ZDV/3TC or TDF/FTC). Use of an integrase inhibitor, or CCR5 antagonist, could be promising, but these agents are not yet recommended owing to insufficient evaluation. As efficacy of PEP is increased with shorter delays of prescription, starter packs of 3–5 days of medication should be available 24 h a day and therefore available in the emergency unit for night and weekend delivery. The exposed person should also receive the necessary support after an accident as this often induces some anxiety, as well as some information about possible adverse events to treatment. An appointment with an HIV specialist is given before the end of the starter pack when PEP was started in the emergency room and/or by a nonexperienced physician. Other treatments

Treatment for HIV

As already stated, and when indicated, PEP should be started as early as possible (the sooner after exposure, the more likely it will interrupt transmission). European and US guidelines allow a window of 48 and 72 h, respectively and New York State’s (USA) guidelines shorten this period to 36 h after exposure [63] . A 28-day course of ARV treatment is recommended, usually with a three-drug combination. Two-drug regimens may be preferred in a case of low-risk exposure and/or for economic reasons (a two-drug regimen is less expensive compared with a full course of three ARVs), and/or for safety reasons, or because adherence is improved by a better tolerability. However, even if it has not been demonstrated that a three-drug combination was superior 436

For women of reproductive capacity who have had exposure to semen, emergency contraception should be discussed. For hepatitis B nonimmune persons (thus mostly after non occupational exposure, as healthcare workers are vaccinated for HBV, the majority therefore being protected), passive immunization with specific immunoglobulins should be considered (within 72 h after exposure) when the source is HBs antigen positive or of unknown status for HBV and completed with hepatitis B vaccination. Serologic screening and follow-up for syphilis are appropriate after sexual contact and treatment if necessary, as well as clinical evaluation of potential other sexually transmitted diseases (e.g., gonorrhea and chlamydia). Expert Rev. Anti Infect. Ther. 9(4), (2011)


Review

Follow-up

Table 4. Recommendations for post-exposure prophylaxis. Follow-up of an exposed individual who Status of source PEP recommended receives PEP is assessed by an infectious Exposure type patient disease specialist or another HIV-care EACS CDC specialist. In addition to clinical followBlood up during treatment (for safety and adher+ Yes (3 drugs) Yes (3 drugs) ence), it is important to monitor liver More severe (intravascular device, HIV deep puncture, visible blood on function, renal function and hematologic Unknown but HIV Yes (3 drugs) To be considered device, or large-bore hollow risk factors (2 drugs) parameters, depending on the medication needle) prescribed. A reasonable follow-up schedHIV No No ule could consist of an early control within + Less severe (superficial injury, HIV Yes (3 drugs) Yes (2 or 3 drugs) 3–5 days of PEP starting, an intermediate solid needle, im. or sc. needle) Unknown but HIV No To be considered visit at mid-therapy (thus 2 weeks of treatrisk factors (2 drugs) ment) and another visit upon completion HIV No No of PEP (4 weeks), all involving clinical + and biological evaluation, adherence sup- Nonintact skin or mucous HIV Yes (3 drugs) Yes (2 or 3 drugs) port and psychological help if indicated. membrane >15 min Unknown but HIV No To be considered Unfortunately, patients are frequently lost risk factors in case of large to follow-up in this setting. A counselvolume exposure (2 drugs) ing intervention, led by specially trained nurses, can significantly improve adherHIVNo No ence to treatment and the proportion of Sexual patients who complete follow-up of HIV HIV+ Yes (3 drugs) Yes testing compared with traditional medical Anal or vaginal sex management (54 vs 23% and 86 vs 54%, Unknown but HIV Yes (3 drugs) Case-by-case risk factors determination respectively) [66] . After PEP completion, all patients HIVNo No should be tested for HIV antibodies at + Receptive oral sex with HIV Yes (3 drugs) Yes 4 weeks and 3 months after the end of ejaculation Unknown but HIV No Case-by-case treatment. Shorter periods of follow-up risk factors determination (6 weeks) are possible only for untreated HIVNo No exposed individuals. Testing for hepatitis B (in noni mmune persons), hepatitis C Intravenous drug use and syphilis (after sexual exposure) and Exchange of syringe, needle, HIV+ Yes (3 drugs) ? a pregnancy test in women (if necessary) preparation material, or any should also be offered. A longer period other material of follow-up is still required for hepatitis EACS: European AIDS Clinical Society; im.: Intramuscular; PEP: Post-exposure prophylaxis; viruses, for instance, 6 months after expo- sc.: Subcutaneous. sure. Biological follow-up is summarized Data from [51,53,101]. in Table 5. Loss to follow-up is probably more frequent in these situations, Exposed individuals should also be advised to avoid blood dona- for example, only 41% of treated women presented to the hospition during follow-up period, until noninfection is confirmed, tal for follow-up in a recent retrospective study [68] . Telephonic as well as to practice protective behaviors with sexual partners or psycho-social support has been offered in a South African study drug-use partners. enrolling 279 rape survivors, which increased the adherence to PEP compared with standard care (from 32 to 38%), but the Care of specific situations difference was not significant [69] . Sexual assault survivors

Unfortunately, sexual assault is relatively common in women and is not uncommon in men [53] . Even if few cases of HIV transmission resulting from sexual assault have been published, there is an increased risk of contamination, as genital trauma frequently occurs and some victims are attacked by multiple assailants. PEP with a three-drug combination has to be offered in such cases, combined with pregnancy prophylaxis in women, particularly as these are not universally available in an emergency care unit [67] . www.expert-reviews.com

Resource-limited countries

Occupational and nonoccupational PEP have now been implemented in resource-limited countries. However, studies on occupational exposure show a low proportion of healthcare workers vaccinated for hepatitis B (51.8% of dental auxiliaries in Nigeria, 51.1% of healthcare workers in a university hospital of Abidjan, Côte d’Ivoire) and a critical need for educational interventions to encourage safe work practices [70,71] . Most PEP are offered 437

Review

Rey

Table 5. Biological follow-up of treated potential exposure to blood or sexual secretions.

Occupational exposure

Nonoccupational exposure

Baseline

Complete blood count ALAT Creatinine (TDF-containing PEP) HIV antibodies HCV antibodies Anti-HBs antibodies†

Complete blood count ALAT Creatinine (TDF-containing PEP) HIV antibodies HCV antibodies Anti-HBs antibodies† Syphilis tests (RPR + TPHA or ELISA)

Day 3–5 (optional)

ALAT ALAT Creatinine (TDF-containing-PEP) Creatinine (TDF-containing PEP)

Week 2

Complete blood count (ZDV-containing PEP) ALAT Creatinine (TDF-containing-PEP)

Complete blood count (ZDV-containing PEP) ALAT Creatinine (TDF-containing PEP)

Week 4

Complete blood count (ZDV-containing PEP) ALAT Creatinine (TDF-containing-PEP) HCV RNA‡

Complete blood count (ZDV-containing PEP) ALAT Creatinine (TDF-containing PEP) HCV RNA‡ Syphilis tests (RPR + TPHA or ELISA)

Children & adolescents

Children and adolescents can be exposed, as adults, by accident with sharp instruments (abandoned syringes for example) but the risk is low and no such contaminations have been reported. Other blood exposures, such as skin contact after fights or playground incidents resulting in bleeding by an HIV-infected child, are at very low risk and usually do not require PEP [53]. PEP is usually more indicated after sexual abuse or nonconstraint sexual exposure. When necessary, specific guidelines have been issued and �� drugs for which pediatric formulations are available might need to be prescribed [79]. For adolescents younger than legal age (