The USPHS document also addresses concerns about unwarranted use of PEP and the use of expanded regimens for exposures that pose a low or negligible risk. For example, data from the Centers for Disease Control and Prevention’s, or CDC’s, National Surveillance System for Health Care Workers, or NaSH, indicate that some HCP take a full course of HIV PEP even when the source is HIV-negative.^7,8 Since 1995, NaSH, a voluntary surveillance system, has collected information on occupational exposures and infections among HCP, including hospital-based dental health care personnel, or DHCP. NaSH monitors routine activities such as immunizations and tuberculosis, or TB, skin testing; occupational exposures to blood, vaccine-preventable diseases and TB; and surveys to evaluate underreporting of exposures. Using a standardized methodology, the CDC can monitor national trends, identify newly emerging hazards for HCP, assess the risk of occupational infection, and evaluate preventive measures. For example, a NaSH study found that more than one-half of exposed HCP took a three-drug PEP regimen when only a two-drug regimen was needed (CDC, unpublished data, 2000).

To date, few data have been published concerning the use of PEP among DHCP.⁹ The objectives of the study whose results are reported in this article were to describe the use of HIV PEP among DHCP included in NaSH. We also discuss the updated USPHS guidelines for managing occupational exposures to HIV and considerations for dentistry.⁴ This article focuses on occupational exposures to HIV specifically, but most exposures also will require evaluation and clinical management for HBV and HCV.⁴

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION SUMMARY REFERENCES

 
The CDC collected data from injury reports among HCP from 52 U.S. hospitals voluntarily enrolled in NaSH. Twenty-two of these hospitals reported exposures among DHCP. We reviewed exposures to blood among DHCP reported from December 1995 through August 2001 to describe characteristics used as surrogates for the quantity of blood to which the worker was exposed; HIV status of the source patient and, for HIV-positive sources, the stage of disease; and the use of PEP by the exposed person, including the number of drugs taken and the duration of treatment. We characterized the quantity of blood to which the person was exposed based on the type of device that caused the injury, whether the device was visibly contaminated with the patient’s blood at the time of the injury, and the depth of the injury. We also examined the types of procedures most frequently associated with the injuries. For hollow-bore devices, the gauge of the needle and purpose (that is, whether it was used to access a vein or artery or was used for blood collection) were identified. An injury was classified as superficial (such as a scratch), moderate (such as penetration of skin) or deep (as indicated by reports such as "muscle contracted," "touched bone").

For each injury for which there was adequate information, we identified the appropriate recommendation for PEP by analyzing the data in two steps. First, we classified the exposure type of each injury as "less severe" or "more severe" by examining the three factors (type of device, visible blood on the device and depth of injury) that characterized the quantity of blood to which the person was exposed (Table 1). Second, we matched the exposure type of each injury with the HIV infection status and stage of disease of the source patient, when known: Class 1 (asymptomatic or known low viral load [ 1,500 RNA copies per milliliter]) or Class 2 (symptomatic, AIDS diagnosis, acute seroconversion or known high viral load [ 1,500 RNA copies/mL]) (Table 2). HIV status either was known at the time of exposure or was determined subsequently by follow-up testing. All HIV tests of source patients tested after the exposure were performed using a standard enzyme immunoassay, or EIA, for which results are not available for at least six hours; data did not indicate when the test results were reported to exposed DHCP. We compiled descriptive data using the Statistical Analysis System (Version 8.2, SAS Institute Inc., Cary, N.C.).

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION SUMMARY REFERENCES

View larger version (27K): [in this window] [in a new window]   Figure 1. Type of device involved in percutaneous injuries (n = 199) among dental health care personnel included in the Centers for Disease Control and Prevention’s National Surveillance System for Healthcare Workers, June 1995 through August 2001.

Injuries were evenly distributed between less and more severe exposure types. Only 62 (31 percent) injury reports contained information about the type of procedure being performed at the time of exposure. Of these, the most common procedures were oral surgical (39 percent), followed by restorative (21 percent), hygiene (15 percent) and those classified as "other" (25 percent). Of injuries for which procedure information was provided, oral surgical procedures accounted for 53 percent (16 of 30) of those with visibly bloody devices, for 80 percent (four of five) of deep injuries and for 53 percent (17 of 32) of injuries categorized as more severe.

One hundred eighty-six source patients were identified. Of these, at follow-up, 24 (13 percent) were HIV-positive, 132 (71 percent) were HIV-negative and 30 (16 percent) had unknown HIV status. About half the injuries in both the HIV-positive group (46 percent) and the HIV-negative and unknown group combined (50 percent) were classified as less severe. Of source patients who were HIV-positive, 25 percent were in Class 1 and 58 percent were in Class 2; this information was not available for 17 percent of these patients (Figure 2). Sixty-three percent of DHCP exposed to an HIV-positive source started PEP, and 33 percent of these people took PEP for 20 or more days. Because of insufficient data, we were unable to determine the actual numbers of PEP drugs used by the exposed DHCP. Of all people exposed to an HIV-positive source, 21 percent warranted use of two drugs and 71 percent warranted use of three drugs, based on the exposure type and the source patient’s stage of disease. (See sidebar, page 1628, for a description of the basic two-drug and expanded three-drug regimens.) Information on the source patient’s stage of disease was not available for 8 percent of exposures that warranted PEP. Twenty-nine DHCP (24 percent) exposed to a source patient subsequently found to be HIV-negative initiated PEP; of 24 for whom information was available, 18 took it for three days or less; five took it for five to 15 days; and one took it for 29 days. In these data, no exposures resulted in HIV infection of DHCP.

View larger version (18K): [in this window] [in a new window]   Figure 2. Human immunodeficiency virus, or HIV, status of known source patients (n = 186) for exposures experienced by dental health care personnel included in the Centers for Disease Control and Prevention’s National Surveillance System for Healthcare Workers, June 1995 through August 2001. HIV-positive, Class 1: Asymptomatic HIV infection or known low viral load ( 1,500 RNA copies/milliliter). HIV-positive, Class 2: Symptomatic HIV infection, AIDS, acute sero-conversion or known high viral load ( 1,500 RNA copies/mL).

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION SUMMARY REFERENCES

Although 50 percent of the injuries in our study were classified as more severe exposures, few were deep injuries or involved large-bore hollow needles or devices used in an artery or a vein. These findings suggest that DHCP were exposed to relatively small volumes of blood and are consistent with the documented low risk of HIV transmission in dental settings. Published reports evaluating percutaneous injuries among DHCP and factors related to the severity of the injury lend further support to the low level of this risk.^13,15–19 For example, cross-sectional and prospective studies conducted by the American Dental Association have found that small-gauge syringe needles and solid instruments caused most (> 96 percent) injuries among dentists; according to current guidelines, these injuries would be classified as less severe.^13,17 In a study examining 428 reports of percutaneous injuries in dental teaching settings, small-bore hollow syringe needles caused 33 percent of injuries, burs 17 percent and explorers 12 percent; only 5 percent were deep exposures.²⁰ In 1995, the circumstances of exposures to HIV-infected blood among 19 dental workers voluntarily enrolled in a national surveillance project commonly known as CDC’s needlestick study were examined.¹⁵ Findings of that study, the first to collect information about factors representative of the amount of blood to which DHCP were exposed, suggested that most of these DHCP were exposed to small volumes. About one-third of the instruments were reported as being visibly bloody at the time of injury, most injuries with hollow-bore syringe needles involved smaller-gauge needles and none of the exposures was described as a deep puncture wound.

In our review of percutaneous injuries to hospital-based DHCP as reported in NaSH, we found that almost half of the devices were visibly contaminated with blood. Where there was information on procedures, we found that slightly more than one-half of the injuries involving a visibly bloody device occurred during oral surgery. We believe, however, that DHCP preferentially reported exposures they believed were more likely to result in HIV infection (those with bloody devices or an HIV-positive source) or for which they wanted PEP or both. This would be consistent with other studies indicating that most percutaneous injuries to HCP go unreported because the person believes the injury carries a low risk.²⁰

The risk of HIV transmission varies with the type and severity of exposure as well as with the source’s HIV status. According to current criteria for recommending PEP, almost three-quarters of percutaneous injuries among NaSH DHCP exposed to an HIV-positive source patient warranted use of the expanded three-drug regimen. This finding was based in large measure on the fact that 58 percent of the HIV-positive source patients were in Class 2 (in which case three-drug PEP always is recommended), and 54 percent of the exposures were considered more severe (again, in which case three-drug PEP always is recommended). We should note that CDC recommendations for PEP are based on the risk of developing HIV infection after different types of exposures, as well as on limited information about the efficacy and toxicity of PEP. Given the low frequency of injuries that were deep punctures or involved large-bore hollow needles or devices used in an artery or a vein, the risk for DHCP in NaSH could well be lower than that for other types of medical HCP, who are more likely to sustain more severe injuries (for example, during blood collection). Understanding the common characteristics of dental injuries as well as the factors associated with the risk of HIV transmission (such as the amount of blood involved) can help the evaluating health care professional balance the risk of HIV infection with the effects of PEP.

Of concern is that some dental health care personnel exposed to source patients who ultimately were determined to be HIV-negative took postexposure prophylaxis.

Of concern is that some DHCP exposed to source patients who ultimately were determined to be HIV-negative took PEP. Two plausible reasons may explain these findings. First, exposed DHCP may have started PEP pending test results of the source patient. If test results had been delayed for some reason, the exposed person may have continued PEP although it was not necessary. This underscores the importance of using the rapid HIV antibody assays and re-evaluating the exposed person within 72 hours so that regimens can be discontinued or altered as additional information becomes available. The use of rapid HIV antibody assays can prevent the unnecessary use of PEP and associated adverse symptoms when results are provided to exposed HCP as soon as they are available.⁸ In addition, the use of rapid HIV test results could result in economic savings.^21,22 In a 1999 study, HCP whose sources were tested using the standard EIA took PEP for a median of four days, with an average cost per person for testing and drugs of $123.²² In comparison, for people whose sources were tested with the rapid HIV test, the duration of PEP was one day, at an average cost of $69 per person for testing and drugs.

A second reason for prolonged PEP use might have been because of concerns that the source was in the "window period" of HIV infection. USPHS guidance, however, indicates that if the source person is HIV-negative and has no clinical evidence of AIDS or symptoms of HIV infection, the likelihood of the person’s being in the "window period" of HIV infection is extremely small. Thus, if PEP is started and the source is found to be HIV-negative, PEP should be discontinued.⁴

Avoiding occupational blood exposures is the primary way to prevent transmission of bloodborne pathogens in health care settings. Reducing percutaneous injuries can be accomplished through engineering controls, such as using safer devices (for example, those with engineered sharps injury prevention features), and by modifying work practices. Personal protective equipment—such as gloves, masks, protective eyewear and gowns—is used to prevent skin and mucous-membrane exposures.

Changes to the bloodborne pathogens standard of the Occupational Safety and Health Administration, or OSHA, mandated by the Needlestick Safety and Prevention Act,²³ were published Jan. 18, 2001, and became effective April 18, 2001.²⁴ The revisions clarified the need for employers to select safer needle devices as they become available and to involve employees in identifying and choosing the devices. Many safer versions of sharp devices used in hospital settings have become available and their impact on reducing injuries has been studied.^25,26 In dentistry, aspirating anesthetic syringes have been developed to incorporate safety features; however, low injury rates in dentistry limit assessment of their effect on reducing injuries among DHCP. Nonetheless, injuries with small-bore hollow needles accounted for the largest percentage of injuries in our study. Furthermore, the impact of safer medical devices in other settings suggests that devices with engineered safety features would reduce percutaneous injuries in dental settings as well. A sharps injury prevention program that includes a process for identifying, screening and evaluating safer dental devices should be developed by all dental practices and integrated into existing infection control and safety programs.

This study has several limitations. It is a retrospective review of self-reported data from less than one-half of the 52 hospitals voluntarily enrolled in NaSH. Thus, the findings may not be generalizable to other hospital-based or private-practice DHCP. In addition, reporting bias may have affected the accuracy of some information, such as the depth of the injury, or it may have resulted in missing data (for example, presence of visible blood on the device, type of procedure during which the injury occurred). Furthermore, loss to follow-up resulted in missing PEP data, which, in many cases, precluded an evaluation of the number of PEP drugs taken. These limitations notwithstanding, however, three important characteristics of the injuries in our study (type of device, depth of injury and presence of visible blood on the device) were similar to those described in several other studies of DHCP.^13,15,17 Finally, these data may represent only a fraction of injuries among DHCP in these facilities, because many injuries go unreported.

	SUMMARY

TOP ABSTRACT METHODS RESULTS DISCUSSION SUMMARY REFERENCES

 
In this analysis, we examined factors associated with the risk of HIV transmission following occupational exposures to blood and examined the use of PEP among hospital-based DHCP enrolled in NaSH. In general, findings of this study are consistent with those of earlier reports indicating that the risk of HIV transmission in dental settings is low. Even so, of the small group of DHCP who were exposed to HIV-positive source patients, close to three-fourths would have warranted the three-drug regimen, based on the severity of the injury and the source patient’s stage of disease. In addition, findings suggest that strategies such as rapid HIV testing of the source patient and follow-up counseling would have decreased the use of PEP among exposed DHCP.

Since the first guidelines for managing occupational blood exposures to HIV were published in 1990, additional information about the use and safety of PEP has become available. Studies have identified factors associated with the risk of HIV transmission, further describing exposures that may benefit most from PEP.²⁷ Improved methodologies for animal and human studies have continued to show beneficial results from PEP in reducing transmission of HIV. Overall, these studies have provided additional support for recommending PEP following certain occupational exposures. With the advent of new antiretroviral agents, however, PEP regimens have become more complex, serious side effects more frequent, and development of resistant strains more common.

Health care professionals who will evaluate exposed DHCP should be selected before DHCP are placed at risk of exposure, should have expertise in antiretroviral therapy and should be familiar with the unique nature of dental injuries so that they can provide appropriate follow-up.

View larger version (121K): [in this window] [in a new window]   Dr. Cleveland is a dental officer, Division of Oral Health, National Center for Chronic Disease and Health Promotion, Centers for Disease Control and Prevention, Mailstop F-10, 4770 Buford Highway, Chamblee, Ga. 30341, e-mail "jlcleveland{at}cdc.gov". Address reprint requests to Dr. Cleveland.