SemOpenAlex |

SemOpenAlex

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2079803492> ?p ?o ?g. }

Showing items 1 to 100 of ±143 with 100 items per page.

W2079803492 endingPage "2105" @default.
W2079803492 startingPage "2095" @default.
W2079803492 abstract "Introduction Early in the AIDS epidemic, there was substantial concern over so-called ‘casual’ or non-sexual transmission of HIV, much of it centered on contact with saliva (sharing an eating utensil or toothbrush) [1,2]. Several epidemiologic studies and national AIDS case surveillance data demonstrated that the transmission of HIV depended on contact with infected body fluids, primarily blood and semen [3,4]. Studies of household transmission of HIV, involving family members who shared commonplace activities, could not demonstrate passage of the virus [5,6]. As the epidemic unfolded, concerns about casual contact abated and widespread major risk factors, such as needle-sharing, anal sexual contact, and transfusion, became the primary epidemiologic focus. Another realignment of emphasis may now be taking place. After about 2 decades, the transmission of HIV in industrialized nations has showed some signs of amelioration. Recent surveillance data in the United States and Europe demonstrate a declining acceleration in the reporting of new incident cases of AIDS [7–9]. Over the past 2 years, there has been a substantial decline in AIDS deaths [7], attributed largely to new therapeutic developments, and continuing reports of declines in risk-taking [10,11]. A number of continuing concerns remain, however. Younger homosexual men may still be evidencing sexual behaviors that place them at risk [12], and the burden of HIV infection in women is growing [9]. Despite continuing concerns about the major risks, overall moderation of the epidemic pace offers an opportunity to reconsider several epidemiologic features whose importance may be modified by the changing epidemiologic picture. Although casual contact with saliva remains an insignificant factor, oral sexual contact may now be of increasing importance in the transmission of HIV. Oral sex may be less efficient than needle-sharing or anal intercourse for the transmission of HIV [13], but its increased use by men who have sex with men (MSM) [14,15] and its prominence in the sexual activity of crack smokers [16,17] may increase its contribution to HIV transmission. In this review, we assess the epidemiologic and anecdotal evidence for oral HIV transmission, review current laboratory investigations of the oral ecology of HIV, and suggest a classification scheme and an epidemiologic approach for further delineation of the role of oral sexual activity. The epidemiologic evidence The first major epidemiologic study of AIDS [18] appeared prior to the recognition of infectious virus in semen, blood, and saliva [19,20] or the availability of serologic testing. In that study, 98% of 50 cases and 99–100% of 120 controls reported receptive oral intercourse (ROI). The authors concluded that there was no difference in orogenital contact between cases and controls. Subsequently, a number of major epidemio-logic studies (including the Centers for Disease Control and Prevention Hepatitis cohort, the Multicenter AIDS Cohort Studies, the San Francisco Men's Health Study) evaluated the risk of oral intercourse, both receptive and insertive, and calculated both adjusted and unadjusted relative risks (Table 1). Most studies calculated the relative risk as the odds ratio, either bivariate or using multiple logistic regression. A variety of study designs and specific risks (e.g., with or without swallowing semen) were used, but results were similar. Despite considerable variation in point estimates (31 out of 39 exceeded 1.0), the confidence intervals (CI) around the relative risk invariably included 1.0 until the mid 1990s (Fig. 1).Fig. 1: . Distribution of relative risks and lower bounds for oral transmission of HIV.Table 1: . Epidemiologic studies measuring oral risk for HIV transmission.Table 1: . ContinuedIn 1993, Samuel et al. [21] studied 83 MSM who had seroconverted and 249 HIV-negative control MSM. The unadjusted relative risk for ROI was 5.3 (95% CI, 2.0–19.0) and for insertive oral intercourse was 3.6 (95% CI, 1.4–13.0). In addition, the relative risk for ROI in the absence of receptive anal intercourse was 3.0 (95% CI, 1.1–12.0). However, the relative risks for insertive oral intercourse in the absence of receptive anal intercourse and for all oral intercourse in adjusted computations were not significantly different from 1.0. These data, suggesting that insertive oral intercourse was not an independent risk factor, did give some credibility to the potential for transmission to those who practice ROI. In 1995, Ostrow and colleagues also found that oral exposure was not an independent risk factor, but noted situations in which it may have occurred, stated that their data were at the limits of detection, and felt uncomfortable with calling oral practices ‘safe sex’ [19–22]. In a large study of 2322 inner-city persons aged 18–29 years, Faruque et al. [16] noted a prevalence odds ratio for oral exposure of 1.5 (95% CI, 1.0–2.1) and an even stronger association with oral sores and ROI (adjusted prevalence odds ratio, 1.9; 95% CI, 1.0–3.6) [16]. Although Wallace and associates did not calculate relative risks, they noted in a study of 3073 female prostitutes that when ROI was the most prevalent sex act, 35.4% of prostitutes were HIV-positive, but when ROI was not most prevalent, 24.2% of prostitutes were HIV-positive (a highly significant difference). Finally, Page-Shafer et al. [23], reporting the results of the Tricontinental Seroconversion Study, found a per partner relative risk for ROI of 1.05, a relative risk for ROI comparing those with 10 oral partners against all others of 1.93, and a relative risk for ROI comparing those with 10 oral partners against those with no receptive anal intercourse of 5.06. All of these results were statistically significant and their 95% CI excluded 1.0 (Table 1). Thus, although earlier studies tended to discount the risk associated with oral intercourse, these more recent investigations provide a substantial basis for reconsideration. Case reports Since 1987, there have been 21 case reports of 42 potential instances of oral transmission (Table 2). Given the thousands of cases of HIV infection that constitute the denominator, such a set of reports would appear to emphasize the rarity rather than the likelihood of the event. The infrequency of case reports may rest, in part, with the difficulty of providing credible validation of the absence of simultaneous risks. In fact, the common thread in these case reports is investigators' considerable efforts to document an exposure other than oral. The rarity of reported oral transmission may be, in part, the result of the rarity with which oral exposure occurs alone, and the hesitancy to assign transmission to oral activity in the presence of simultaneous risks. However, the fact that it appears to confer greater risk when it is the dominant (although not exclusive) behavior [24] gives impetus to the task of sorting out the contribution of oral intercourse to transmission.Table 2: . Case reports and observations on oral transmission.Transmission estimates Perhaps because of the perceived inefficiency of transmission, little attention has focused on the probability of transmission through oral intercourse, and it is not specifically addressed by several recent reviews [13,25]. The European study of heterosexual partners identified 50 couples who practiced unprotected orogenital sex, but protected penile-vaginal and penile-anal sex, and detected no HIV transmission after a median of 24 months of observations [26]. The authors noted, however, that with the small number of observations, the 95% CI around the estimate of transmission was 0.0–4.3 per 100 person-years. In addition, the longterm stability of these relationships suggests that the infected partner may have been in a stable stage of HIV progression, during which infectivity may be low. In an innovative approach, Samuel et al. [21] used the San Francisco Men's Health Study to provide data-based estimates of HIV prevalence, sexual intercourse activity, and seroconversion status and used them to model the remaining unknowns (infectivity parameters). Based on several different models, the per-partner infectivity of ROI was about 1% (range, 0.85–2.3%), where ‘per-partner’ refers to the risk with a given partner, uncontrolled for activity level, and should be distinguished from the risk of a single sex act (‘per-contact’ risk). In comparison, the per-partner infectivity of anal receptive intercourse was about 10% (range, 4.2–12%). In a modeling study that incorporated the stage of infection, it was estimated that oral sex acts imposed a transmission risk that was one-sixth that of anal sex acts [27]. In an assessment of per-contact risk for transmission associated with four types of homosexual contact, Vittinghoff et al. [28] estimated that the risk for unprotected receptive anal sex (0.24%; 95% CI, 0.05–0.43) was eight times the risk for unprotected receptive oral sex (0.03%; 95% CI, 0.01–0.18). Other modeling endeavors have selected arbitrary parameters for transmission and focused on the subsequent dynamics. For example, Koopman et al. [29] modeled the number of new AIDS cases that would occur in a simulated epidemic involving oral and anal preference groups, and used an estimate of transmission per oral sex act of between 0.0005 and 0.01, setting these at one-quarter the corresponding estimates for transmission per anal sex act. Biologic plausibility The biologic risk for transmission or acquisition of HIV from oral sexual contact is not known, but the risk is likely to be related to a number of factors. These include the presence or absence of virus at sexual sites (oral, vaginal, anal and penile), the titer of virus (if present), the integrity and mechanical properties of the sexual mucosa, mucosal immunity, local inhibitory factors, and the presence or absence of cofactors that may facilitate transmission. Finally, the frequency and nature of exposures (e.g., the relative effect of a large number of lower risk events compared with a small number of higher risk events) and the underlying epidemiologic features of HIV dynamics in the community may have an impact on the frequency of HIV transmission from oral intercourse. Assessment of many of these factors has only recently been undertaken. Evidence from animal experiments There is recent animal experimental evidence for and against ROI as a route for transmission. Stahl-Hennig and colleagues applied small amounts of simian immunodeficiency virus (SIV) to tonsils of rhesus monkeys and demonstrated that the tonsils may be a portal for non-traumatic entry of SIV [30]. Similarly, Baba et al. [31] were able to infect adult macaques in the absence of oral lesions with doses considerably below those required for infection through rectal exposure. In contrast, Bosch et al. [32] exposed 11 Macaque nemestrina neonates to HIV-1 infection via the oral, intravenous and rectal routes. Three out of five exposed rectally, two out of two exposed intravenously, and none out of four exposed orally became infected. The authors pointed to potential differences between SIV and HIV, and stated that the latter was not efficiently transmitted in macaques. Nonetheless, the possibility of atraumatic oral infection heightens the urgency of determining the mechanisms of action of inhibitors, and their role in vivo. Infectious HIV at sexual sites It is now well established that infectious HIV can be found in semen, as well as in cervical and vaginal secretions. The potential for HIV transmission from these sites to oral mucosa has not been well studied, although inhibitory substances in saliva (see below) suggest that such transmission is not facile. More recently, information has emerged about the presence of infectious HIV in pre-ejaculatory fluid. Pudney and colleagues reported the presence of HIV-1-positive cells in three sperm-free samples from four donors [33]. They also noted a variable number of positive pre-ejaculate specimens from persons who were symptom-free (five out of six), symptomatic (one out of three), on zidovudine (two out of five), and not on therapy (four out of four). The presence of virus in these situations highlights the possibility of penile-oral transmission of HIV even in the absence of ejaculation. Similar possibilities would exist from the transfer of vaginal/cervical secretions, and would similarly be dependent on a variety of as yet incompletely explored factors that affect the degree of shedding and the receptivity of the oral cavity to infection. Virus in saliva As noted above, the first report of virus in saliva [19] was followed quickly by a more extensive evaluation that found virus in only one out of 83 specimens [34]. Subsequently, investigators have reported a frequency of detection from 0 to 83%, depending on the type of specimen and the type of laboratory method (Table 3), but most of this work occurred before current staging methods were available. Culture attempts have generally placed the isolation proportion at under 5%, with the exception of a 21% positivity for low levels of virus found by Yeung et al. [35]. Results with PCR testing have been more variable. For example, in an extensive study, 218 simultaneous blood and saliva specimens from 75 HIV-positive persons were tested for viral p24 antigen and infectious virus; 38% of blood specimens and 1% of saliva specimens were positive for cell-free infectious virus [36]. Goto et al. [37] used PCR testing to study 20 patients for long terminal repeat (LTR), gag, and env proviral sequences. With LTR probes, 10 out of 20 specimens were positive, but only 25% were positive with probes for gag and env. Repeated testing confirmed the higher positivity for LTR. With PCR testing, Quereshi and colleagues detected virus in 83% of specimens [38]. In an extensive review of the mechanisms of infectivity for salivary secretions, Shine et al. [39] concluded that the precise balance between infectious secretions and inhibitory effects in vivo remains to be elucidated. They stated that, ‘it is important to establish the correlations among the amount of HIV, its infectivity in salivary secretions, and the level of anti-HIV factors present in these secretions’ [39].Table 3: . Frequency of detection of HIV-1 in saliva.Inhibition by saliva Coincident with early reports suggesting low level of HIV in saliva, early reports also determined that there were components in saliva that may inactivate HIV [40]. Subsequent research has clearly established the presence of such substances [39,41,42]. A number of mechanisms have been postulated for this inhibitory activity, including cell disruption, soluble antibody, high molecular weight proteins, and soluble anti-HIV components. Baron and Cloyd [43] proposed that HIV transmission is inhibited by saliva-induced disruption of infected cells and neutralization of maturing HIV in cells made permeable by saliva. They demonstrated 30–45% cell disruption by saliva. Sanneman et al. [44] suggested that immune factors in whole saliva of HIV-1-infected persons, such as HIV-1-specific IgG from serum, may produce inhibition. As reviewed by Shine et al. [39], submandibular, submaxillary and parotid secretions, as well as whole saliva, reduce HIV infectivity in vitro and several modes of inhibition are discussed. High molecular weight salivary proteins such as mucins may entrap HIV-1 [45,46], although this mechanism has not been fully explained because anti-HIV-1 activity is retained in human parotid saliva, which does not contain mucin. Second, there may be a role for soluble anti-HIV components. Although a host of soluble proteins has been identified, and several (cystatin, lysozyme and lactoferrin) have anti-HIV-1 activity at concentrations above those found physiologically, the protein of greatest interest is recombinant secretory leukocyte protease inhibitor (SLPI). The mechanism for inhibition is under intense investigation. McNeeley and coworkers [47] observed that in vitro infection of monocytes by HIV-1, even after only 1 h of exposure to saliva, as measured by viral reverse transcriptase activity, is suppressed for 3 weeks after infection. They found that SLPI was inhibitory at physiologic concentrations and that the anti-HIV-1 antibody activity was dose-dependent. SLPI was shown to target a cell-associated molecule, but its inhibitory activity did not involve direct interaction with CD4 antigen. It has been suggested that SLPI's mode of action may be to inhibit the virus-cell fusion or uncoating step in the HIV infection cycle [48]. However, Shine et al. [39] present evidence that SLPI may inhibit virus-cell fusion but does not inhibit cell-cell fusion. A potential critical observation was made by Wahl and colleagues [49], who noted that SLPI accumulated in acinar cells or ductal epithelium, and that HIV-1 transcripts did not. The latter were found only in mononuclear cells within salivary gland stroma (in 19 out of 55 specimens tested), and there was no evidence of productive HIV-1 infection in salivary gland epithelium. This lack of colocalization suggests that SLPI may function to prevent transmission and infection, rather than influence established infection. From an epidemiologic point of view, it may operate to prevent acquisition by someone performing ROI, but may not play a role in preventing transmission from an infected person to a partner who practices oral insertion. Antibody in saliva The presence of anti-HIV antibody in saliva has been well recognized for over 10 years. As summarized by Malamud and Friedman [41], 15 studies demonstrated a concordance of 564 (97%) out of 582 paired specimens of saliva and serum. This correlation holds for whole saliva as well as for parotid saliva, and there is a virtual absence of false positives [50,51]. These initial observations have formed the basis for oral testing for HIV. Evaluation of available oral tests (14 studies) reveal a sensitivity of 97.2–100% and a specificity of 97.7–100% [52,53] as well as improved safety, ease of collection, and patient acceptability. Similar results were achieved under field conditions in a study of MSM in Mexico City [54]. The sensitivity, calculated from 369 paired saliva-serum samples from a low prevalence population and 134 paired specimens from HIV-positive persons, was 98.5%, and the specificity was 100%. Although anti-HIV antibodies are an invaluable diagnostic aid, their role in the oral transmission of HIV is not yet clarified [39]. The role of oral pathology The numerous ulcerative and non-ulcerative conditions that affect the oral cavity [55–60] may alter the biologic activity of infectious HIV and of inhibitors. In a review of periodontal complications, Mealey [55] noted that linear gingival erythema, a condition characterized by unresponsiveness to local debridement and mechanical plaque control, and necrotizing ulcerative periodontitis (NUP), characterized by rapid destruction of alveolar bone, play an important role in revealing to the clinician the overall clinical pace of HIV infection. NUP, for example, is a good predictor of CD4+ cell counts of below 200×106/l, and in one study was a strong predictor of rapid mortality [61]. The true prevalence of these conditions is difficult to assess because of the bias of ascertainment when clinic populations are used. In Mealey's review of existing studies, the prevalence for NUP was generally below 10%, although several studies report much higher values [55]. When considering the pathology of the entire oral cavity, a variety of neoplastic (Kaposi's sarcoma, non-Hodgkin's lymphoma), viral [herpes, human papillomavirus, Epstein-Barr virus (hairy leukoplakia)], bacterial (anaerobes, facultative anaerobes, Mycobacterium avium complex), and fungal (candidiasis, histoplasmosis) conditions are well described [58] as well as a variety of other conditions [epithelioid angiomatosis (Rochalimea spp.), apthous-like ulcers] in which infection plays a role [58]. Shiboski and colleagues [62] conducted a prospective study of oral lesions and HIV conditions on 176 HIV-infected women and 117 HIV-negative women who were at risk for infection. The prevalence of oral lesions was significantly higher amongst infected women (22%) than uninfected women (3%), and there was a substantial association between the presence of oral lesion and a CD4+ cell count of below 200 × 106/l (OR, 8.9; 95% CI, 2.6–30.0) [62]. the authors noted that amongst those with CD4+ cell counts below 200 × 106/l, hairy leukoplakia was 5.5 times more common in those infected heterosexually than amongst injecting drug users. Maternal-infant transmission through breastfeeding Transmission of HIV by breastfeeding is well established, but the precise mechanism for transmission by this route is not known. Unlike saliva, levels of HIV are high in breastmilk. Although SLPI levels are high in both colostrum and breastmilk, as well as the saliva of the newborn immediately postpartum, only breastmilk shows a dramatic decrease in SLPI levels over the few weeks postpartum. The potential relationship of this decline to HIV transmission via breastmilk requires further elucidation, particularly in relation to other potential risk factors such as breast abscess, nipple trauma, and oral lesions in the infant. Several investigators have tried to understand the role of saliva, SLPI, and oral factors in maternal-infant transmission. Southern and Southern [63] observed that productively infected milk monocytes and macrophages can be detected by in situ RNA hybridization and immunocytochemistry in the breastmilk of HIV-1-infected mothers. Adhesion molecules in infant saliva may be important in transporting infected cells into tonsillar and intestinal crypts, and may facilitate HIV transmission during breastfeeding. Janoff et al. [64] found that levels of SLPI were well below the HIV-1 inhibitory level in breastmilk from nursing mothers from Rwanda and the United States. There was, however, an elevation of SLPI in colostrum. These observations support their epidemiologic counterparts that oral transmission may be low, but maternal-child transmission by breastmilk is appreciable [49]. Discussion The epidemiologic and biologic features of HIV transmission undergo continuing reevaluation. In the case of oral transmission, dismissal in the early years of the epidemic [65] was replaced by anecdotal recognition of its possibility (Table 2), and more recently by studies which suggest that oral sex, especially ROI, is an independent risk factor for transmission (Table 1). There are several potential reasons for the discrepancy between early findings and more recent ones. First, earlier studies dealt almost exclusively with MSM, whose frequency and intensity of receptive anal intercourse may well have obscured any small effect of oral intercourse. Second, the recent studies of MSM [22,23] carefully delineated the time of seroconversion, and were thus dealing with sexual behavior that may have been more directly related to the actual transfer of virus. Third, these later studies have taken place at a time of changing sexual practices as well as increased HIV prevalence. As noted, the decline in unprotected receptive anal intercourse and the possibility that MSM are favoring ROI gives the latter's potential for transmission more prominence, whatever its actual magnitude [14,15]. Finally, the other studies that indicate the potential for oral transmission [16,24] were conducted in environments in which transmission by MSM does not predominate. Rather, young, inner-city, minority persons and prostitutes, whose risks are more linked to drug use and heterosexual activity, were the primary focus. In such circumstances, receptive anal intercourse, although present, is less intense, oral intercourse may play a larger role, and its role may be more amenable to discovery. Thus, the reevaluation of the role of oral transmission rests on a confluence of events that includes a decline in the risks that appear to be most effective in propagating the virus (e.g., infected blood products, anal sexual activity, needle-sharing), an increase in oral sexual activity, both among MSM and among persons who smoke crack cocaine, oral pathology that is associated with HIV infection, and oral lesions that may result directly from smoking crack cocaine. To these can be added the poor state of oral health of inner-city persons who use drugs, work as prostitutes, and are involved in the oral sexual activity that is a concomitant of crack cocaine use (unpublished observations). This confluence raises several issues of public health importance. The presence of chronic conditions, the occurrence of chronic ulcerating lesions (candidiasis, herpes simplex virus infection, apthous ulceration, ulcers secondary to crack cocaine use), and the presence of many oral pathogens may provide an opportunity for facilitation of HIV transmission similar to that which occurs with sexually transmitted diseases [66] (J. Wasserheit, personal communication, 1998). In the genital tract, these conditions and pathogens create an environment of increased HIV shedding, and elevate the relative risk for HIV transmission by three-to fivefold. If such an analogy exists for the oral cavity, it is possible that such increased shedding could overwhelm the natural inhibitory activity of saliva and other oral secretions, and would contribute to the circumstances that heighten the importance of oral transmission. Just as treatment of sexually transmitted diseases in persons with and at risk for HIV may have an ameliorative effect on transmission, the aggressive treatment of oral lesions and conditions may reduce the potential for oral sexual transmission of HIV. The need for more definitive epidemiologic and biologic evaluation is heightened by the problems of providing authoritative counseling to persons about the risks of HIV and oral sex. Currently, there is considerable ambiguity about what can be said, based on evidence, with regard to what is ‘safe’ and what is ‘safer’ [67]. Counseling guidelines are hotly debated, not only because of differing interpretations of the available data, but because of different approaches to how counseling should be conducted [68–71]. The recent introduction of antiviral therapy and viral load therapy has further complicated people's perception of risk and the tailoring of messages [72]. Some of the confusion may relate to a failure to clarify exactly what is meant by oral sex, a failure to distinguish between transmission and acquisition, and a neglect of the setting in which the sex act takes place. In theory, oral sex could encompass insertive and receptive combinations of the oral cavity with four potential sexual sites: oral cavity, vagina, rectum, and penis. All eight combinations permit the exchange of saliva but only one, receptive penile-oral intercourse, may expose the receptor to semen. Current evidence would suggest that the risk of HIV transmission from exposure to saliva is considerably smaller than the risk from exposure to semen [13,25]. Thus, exposure to semen during the sex act probably imparts a heightened risk, and other oral sex acts may also differ with regard to the probability of HIV transmission. A rational approach to counseling would require better information on the nature of these risks, how they are affected by the oral milieu, and by local epidemiologic circumstances, and, by analogy with vaginal shedding of HIV [73], with plasma viral titer. Similarly, the proportional importance of oral sex to HIV transmission will be a complex result of the relative frequency of oral sex compared with other activities, infectivity of oral secretions and its modification by oral pathology, resistance to infection by inhibitory substances in saliva, the HIV prevalence in the community in which such activity takes place, the maturity of the epidemic in the community (given recent observations on differential infectivity by stage of infection [74,75]), the role of high activity antiretroviral therapy, and the extent to which personal prophylaxis is adopted. A research agenda that clarifies the risk from oral sexual contact is certainly important for rational counseling and prevention, but even a firm estimate for transmission may not be adequate. The evidence from this review supports the notion that unprotected ROI is of low but non-trivial infectivity and widespread occurrence. It may be misleading to couch future recommendations in terms such as ‘safe’ and ‘safer', which are ambiguous and subjective at best, since risk is a function of both the estimate for transmission and the frequency of the event. Perhaps recommendations and counseling that incorporate the importance of persons’ awareness of their own risk orientation in personal decision-making can lead to a better appreciation of the spectrum of ‘safety’." @default.
W2079803492 created "2016-06-24" @default.
W2079803492 creator A5002511362 @default.
W2079803492 creator A5016390987 @default.
W2079803492 creator A5033359726 @default.
W2079803492 creator A5046491812 @default.
W2079803492 creator A5073223457 @default.
W2079803492 date "1998-11-01" @default.
W2079803492 modified "2023-10-18" @default.
W2079803492 title "Oral transmission of HIV" @default.
W2079803492 cites W1910394161 @default.
W2079803492 cites W1910840782 @default.
W2079803492 cites W1935830129 @default.
W2079803492 cites W1966991108 @default.
W2079803492 cites W1971965350 @default.
W2079803492 cites W1980874481 @default.
W2079803492 cites W1984620851 @default.
W2079803492 cites W1987761869 @default.
W2079803492 cites W1988373208 @default.
W2079803492 cites W1991227862 @default.
W2079803492 cites W1994117224 @default.
W2079803492 cites W1996849923 @default.
W2079803492 cites W1999141010 @default.
W2079803492 cites W2000181247 @default.
W2079803492 cites W2000389746 @default.
W2079803492 cites W2002223086 @default.
W2079803492 cites W2003224210 @default.
W2079803492 cites W2003358027 @default.
W2079803492 cites W2004363016 @default.
W2079803492 cites W2008438368 @default.
W2079803492 cites W2013788654 @default.
W2079803492 cites W2013926797 @default.
W2079803492 cites W2017207315 @default.
W2079803492 cites W2019081608 @default.
W2079803492 cites W2021981648 @default.
W2079803492 cites W2026834471 @default.
W2079803492 cites W2030784485 @default.
W2079803492 cites W2035752457 @default.
W2079803492 cites W2036542823 @default.
W2079803492 cites W2047318471 @default.
W2079803492 cites W2050350607 @default.
W2079803492 cites W2052415071 @default.
W2079803492 cites W2052980876 @default.
W2079803492 cites W2054684353 @default.
W2079803492 cites W2054778547 @default.
W2079803492 cites W2058075938 @default.
W2079803492 cites W2059005571 @default.
W2079803492 cites W2061769247 @default.
W2079803492 cites W2065249995 @default.
W2079803492 cites W2065415038 @default.
W2079803492 cites W2070236534 @default.
W2079803492 cites W2074086126 @default.
W2079803492 cites W2075136384 @default.
W2079803492 cites W2080431531 @default.
W2079803492 cites W2081861819 @default.
W2079803492 cites W2082250341 @default.
W2079803492 cites W2087038250 @default.
W2079803492 cites W2089813082 @default.
W2079803492 cites W2091204458 @default.
W2079803492 cites W2092153132 @default.
W2079803492 cites W2093927991 @default.
W2079803492 cites W2124767354 @default.
W2079803492 cites W2136088948 @default.
W2079803492 cites W2137425191 @default.
W2079803492 cites W2139053866 @default.
W2079803492 cites W2147206752 @default.
W2079803492 cites W2148000796 @default.
W2079803492 cites W2161772098 @default.
W2079803492 cites W2172593462 @default.
W2079803492 cites W2243021045 @default.
W2079803492 cites W2256147900 @default.
W2079803492 cites W2258047503 @default.
W2079803492 cites W2275610698 @default.
W2079803492 cites W2312778470 @default.
W2079803492 cites W2323866244 @default.
W2079803492 cites W2324817321 @default.
W2079803492 cites W2327441472 @default.
W2079803492 cites W2341804666 @default.
W2079803492 cites W2419199114 @default.
W2079803492 cites W2441991313 @default.
W2079803492 cites W2615187673 @default.
W2079803492 cites W4240502965 @default.
W2079803492 cites W4242669616 @default.
W2079803492 cites W4249185302 @default.
W2079803492 cites W4250505167 @default.
W2079803492 cites W4251413849 @default.
W2079803492 doi "https://doi.org/10.1097/00002030-199816000-00004" @default.
W2079803492 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9833850" @default.
W2079803492 hasPublicationYear "1998" @default.
W2079803492 type Work @default.
W2079803492 sameAs 2079803492 @default.
W2079803492 citedByCount "133" @default.
W2079803492 countsByYear W20798034922012 @default.
W2079803492 countsByYear W20798034922013 @default.
W2079803492 countsByYear W20798034922014 @default.
W2079803492 countsByYear W20798034922015 @default.
W2079803492 countsByYear W20798034922016 @default.
W2079803492 countsByYear W20798034922017 @default.