Tobacco Documents Online

ENVIRONMENTAL TOBACCO SMOKE: A COMPENDIUM OF TECHNICAL INFORMATION Technical Review - Michae). W. nrrden, Ph.^= Chapters 5 ("Measuring Exposure to Environnu.ntal Tobacco Smoke"), 6 ("Exposures to Air Pollutants"), and 7("]sxposure Assessment in Passive Smoking") all deal with essentia:lly the same subject matter. Although the approaches taken represent the various author's individual viewpoints, it would be in the best interest of this Compendium to combine these three chapters into one comprehensive review of "Exposure to Environmental Tobacco Smoke". This reviewer will address each chapter individually with emphasis on omissions, redundancies, and cnntradictio:«s. - Chapter 5 - "Measuring Exposure to Environmontal Tobacco Sraokell Chapter 5 is in a form which makes it 3ifficult to review adequately. Specifically, there are numerous misspellings, incomplete or awkward sentences, and the list of literature references is unavailable. The subject matter of biomarkers is not really.addressed at all. The limitations of nicotine and RSP are not adequately portrayed. There are ways to make RSP a much more useful marker of ETS: through the use of apportionment techniques such a~= t::e published methods tor UVPM (ultraviolet particulate * matter, references" below) and solanesol ( j73 ]..and references below) and the presented but not yet published method for FPM (fluorescent * particulate matter) [751. At a minimum, the pul:)lished methods must *"[]" refers to Chaps. 5-8 Index.

6 t. U11 be included. p. 53 - The opening sentence is misleading. It st?*_p-- that measuring exposure to ETS is an exact science when in fact it can only be estimated. Replace "assessing" with "estimating". p. 54 - The author states that "models can be developed. and validated to predict concentration" implying that this has been done. This reviewer knows of no adequately validated models for this purpose, including those referenced in Chapter 7. p. 55 - The opening sentence is misleading. It states that "(ETS) is -a complex mix of over 4,000 air contaminantF". This is speculation based on what is known about mainstream and sidestream tobacco smoke. Nowhere near 4,000 tobacco smcke derived components have ever been measured at true ETS concentrations. In fact, this reviewer is unaware of more than 100 or so. p. 56 - Although RSP and nicotine are widely used as markers, this chapter (indeed, all three chapters) neads to address the appropriateness of their use. The author states they are used because of "their relationship to other ETS contaminants". This reviewer is unclear as to what is meant by this statement, when in fact,-the relationship between nicotine and PSP is highly variable across environments. Tb i c section needs eiaboration with appropriate documentation. p. 57 - "The EPA standard is for particle mass, 10 µm." What does 2

i [l this mean? Ten µm is a unit of particle diameter, not mass. The author needs to clarify what is meant by "reasonably consistent RSP to vapor ph%~kse ratio(*:" . _`or ETS" and "nicotine, for some applications, may vary with ETS related RSP". While nicotine and RSP are almost always correlated in field studies (i.e., there is a statistically significant slope), the corre].ation (i, e. , goodness of fit) is almost always very poor. This reviewer questions how this supports the statement that nicotine ca:ri be used to estimate RSP attributable to ETS. p. 58 - The author incorrectly states that there are no health standards for exposure to nicotine. 0!SHA has one (500 µg/m3). A further drawbac;t ot nicotine as an ETS m<<rker that should be included is the fact that in environments w:Cthout active smoking activity (but which have a history of smokiiig activity) there is the very real possibility of a nicotine background due to * adsorption/desorption from room furnishings, etc. t74}. In these cases, measurement of nicotine indicates expo:aure to nicotine only, and not ETS (other less-adsorptive components have decayed to zero). The author raises this issue at the bottom of the page and states that "background levels of nicotine might also be indicative of outgassing from surfaces of other volatile ETS components". This latter statement is purely speculation <<nd should be omitted. p. 61 - In the discussion of short-term me,==ement methods for ETS RSP, the ultraviolet absorption method reported by Ingebrethsen * et al. [49] needs to be included. This UV method is the only methodology widely-used in field surveys i.o indicate some ETS *"[]" refers to Chaps. 5-8 Index. 3

* apportionment of RSP [.15,70,75,92].- rl The author states incorrectly that "}hic tXA.D-4; method has been evaluated for sampling periods up to one hour with...detection limit of 0.1 µg/m3". The method has been evaluated for periods up to 8 hours with LOD of 0.02 µg/m3. LOD for 1 hour sampling is 0.17 µg/m3 [77]* pp. 61-62 - It is not true that particle phase nicotine can be determined by analyzing the first filter. Particle phase nicotine will impact the first filter but is stripped from the particles and collected on the second filter as vapor phase nicotine. This method cannot give separate i:.forinaiion on vapor phase and-particle phase nicotine; it can only give total ETS nicotine if both filters are analyzed and the results summed (i.e., the first filter underestimates particle phase nicotine and the second filter overestimates vapor phase nicotine). Although passive samplers are useful in field surveys under certain constraints, they generally show much worse ;precision than active sampling systems and, in some cases, severely overestimate nicotine concentrations [76]*. p• 63.- Delete the gratuitous statement that "ETS is a complex mix of ~pveral thousand chenicals..." for the raasons outlined comment to p. 55. The author states that RSP and vapor phase nicotine * "[]" refers to Chaps. 5-8 Index. 4 are

r "reasonable" markers. How is reasonable defined? This is solely the authors opinion and should be stated as such, if not deleted. Chapter 6 - "Exposures to Air Po].lutants's The title for Chapter 6 listed in the Table of Contents does not agree with the actual chapter title. It appears the title listed on p. 65 is more indicative of the chapter contents and implies that this chapter was actually written for some other purpose. This is further supported by the numerous irrelevant comments attributable to air pollution in general. As stated previously, this chapter should be combined with Chapters 5 and 7 for a more comprehensive and useful review. pp. 66-67 - What is-the relevance of the information relating to sulfur, zinc, bromide, lead, silicon, calcium, and iron? should be deleted. This The definition of mainstream smoke is circuitous and needs revision. This reviewer disagrees with the authors' statement that exhaled mainstream tobacco smoke adds little to ETS. Exhaled mainstream smoke can easily aa:' 1c? -20°s to true ETS In addition, the chemical composition of exhaled mainstrean smoke is markedly different from sidestream smoke due to absorption in the smoker. 5

i t- L These authors have included the same gratuitous comments regarding the "several thousand chemical compounds" in ETS as was included in Chapter 5_ These c.om?aents should be del eted for the reasons outlined previously (comments to pp. 55 & 63). p. 68 - The authors state that the "large numibers of constituents in ETS make it impossible to assess overall exposure based on...each one" when in fact this is not true. The reason exposure is not assessed on each constituent is because the vast majority of tobacco-smoke chemicals have never been detected in ETS! In fact, validated analytical methods exist for only a very few. Although this may seem a trivial pursuit, the authors are implying "ovPn•::elming ccrupiexity" is the relevant issue when in realit~, "extremely dilute concentration" is more pertinent. The authors should strive to make a more objective presentation of the issues. Of the possible measures of ETS, why do the authors choose RSP, nicotine, cadmium and nitrosamines? RSP and nicotine form the major theme of Chapters 5 and 7. Cadmium and nitrosamines are not generally useful ETS tracers because of the ultra-trace quantities present. (Is this a scare tactic?) Cadmium and nitrosamine related information should be deleted and the remainder with Chapters 5 and 7. integrated p. 69 - The statement is made that "each --:oker in the home raised the mean respirable particle level by 20 lag/m3". What is the relevance of this increase? In relation to 'the OSHA standard for dust of 2000 Ecg/m3, this increase seems a.inuscule. The data 6

i i I+ ii I presented in the next paragraph (0.88 and 2.11 µg/m3) are considerably lower. This reviewer fails to see the relevance or significance of these dat?.. M^ro information (with references) needs to be presented or this discussion eliminated. Because of the fact that RSP is not selective for ETS, it is impossible to infer that the 962 µg/m3 measured (or any of the other data) is attributable to ETS. p. 71 - The authors contradict themselves by stating that "Indoor cadmium levels were below the detectable limit" followed by "cadmium levels were highly correlated with...particulate measurements". If 7ad:uium was undetected, how could it be correlated with anything else? The relevance of any discussion of cadmium data is unclear and should be omitted. The relevance of the information presented on nitrosamines is unclear and appears to be taken out of context [9]. While true that Brunnemann et al. detected nitrosamines, they were not directly attributable to ETS (i.e., they were not tobacco-specific nitrosamines). Brunnemann et al. state that "...the assessment of traces of established animal or human carcinogens, possible thrqugh advanced instrumentation, does not imply increased cancer risk for man. _ The human risk can be established only through appropriate epidemiological investigations." They go on to state =~...no Go epidemiological data exist linking human respiratory cancers to pp ~ volatile nitrosamines." The entire discussion of nitrosamines N should be omitted from this chapter and left to the experts ~ rV 7

L.J r ~ ~ (Hoffmann et al., Chapter 8). What is the relersnce of the breath concen' r=* i c:: of benzene in i! smokers? This paragraph must be deleted. pp. 71-72 - What appear to be trivial increases in RSP (29 to 56 µg/m3) and nicotine (0.3 to 2.5 µg/m3) values; are presented. What is the significance of these increases? Likewise, the statement is made that "exposu::-e was increased by 20 llg/m3" • Z ~N What is the relative magnitude and significance of this increase? . The two paragraphs presented on cotinine measures are hardly adequate. They should be deleted or covered in much greater detail. Summary statement #2 declares that ETS is a substantial contributor to indoor air of benzene, acrolein, nitrusamines, pyrene, and carbon monoxide. With the exception of bens,ene and nitrosamines, the other compounds are not discussed within this chapter. The data presented for benzene relate only to smokers and the nitrosamine data are not specific to ETS. 7'his summary statement is unfounded, unwarranted, and should be de].eted. 8

Chapter 7-O'Exposure Assessment in Passive Smoking" The 1.Yt5!rature cita_tions for Chapter 7 are incomplete (e.g., IA.RC, 1987; Williams, 1985). The frequent use of "persortal communication" citations is inappropriate. If this information is not included in the scientific literature or other public domain, it must be deleted. Considering the scope and content of Chapters 5 and 6, this reviewer sees no real need for this information to be presented as a separate chapter. What new information is presented here should be incorporated with Chapters 5 and 6 for a more comprehensive and cohesive review. p• 7y - This reviewer agrees with the author that exposures tu ETs can be assessed by personal air contaminant zaonitoring (presuming appropriate contaminants are monitored); however, this reviewer strongly disagrees that exposures can be assessed by modeling. This is a common misstatement in Chapters 5, 6, and 7: assessment implies an actual determination. Modeling can, at best, provide an estimate, i.e., in place of an actual measurement. In reality, modeling is still only a crude estimate and needs to be portrayed as such. p. 80 - It is not true that the "two most proa.ising markers for ETS are respirable suspended particles...and nicotine". While these are currently the most widely uscd, :aany resaarchers in the field are actively pursuing identification of other markers specifically to overcome the significant-limitations of-taiese two. i 9

The author offers several unsubstantiated opinions as fact: "the substantial emission of RSP", "ETS is the daminant contributor", "RSP...found to be subc}antially elevated", etc. If this Compendium is to be a scientific document, these loosely defined qualifiers (substantial, dominant, etc.) need replaced with quantifiers and appropriate literature citat:.ons added. kJ I L L*~ U The author states that the presence and number of smokers lacks specificity for tobacco smoke. What does this mean? Assuming the author means that RSP lacks specificity for tobacco smoke, this is a true statement. However, the author continues throughout the rest of the chapter blatantly ignoring this :aevere limitation. Again, the author incorrectly states that "RSP is currently the best...to represent ETS". This is opinion and should be stated as such, if not deleted. The author states that "nicotine and ...cotinine...derive exclusively from tobacco products". This is incorrect. Nicotine has been determined in a substantial number of foodstuffs including tomato, potato, eggplant, pepper, and instant tea [13,100]* . This reviewer agrees that tobacco smoke*is the most important source; however, the possibility for nicotine and co-tinine in body fluids arising from dietary sources cannot be ignored. It is unclear how the statement "Generally, the mean concentrations c:fi -'~cotine and cotinine in...nonsmokers exposed to ETS under natural conditions" can be used to support the claim that "smokers are present in nearly all environments". This section needs rewritten. * "[]" refers to Chaps. 5-8 Index. 10

The assumption (statement) made in the last paragraph on p. 80 (and later in the chapter) that "the averaRe...=!*cyer...emit about 22 mg of RSP per cigarette" is totally unjust:Lfied. This reviewer must assume that the IARC reference gives an average sidestream 'tar' of 22 mg/cigt (which is not unreasonable). How is it possible to justify the equation that 22 mg SS 'tar' = 22 mg ETS RSP? Even the author recognizes the fact that "almost all nicotine shifts from the particulate phase in MS and fresh SS smoke to the vapor phase in ETS". It must be assumed thai: this same phenomenon occurs for other tobacco-smoke constituents. The weight of ETS RSP must be substantially less than the 22 mg predicted from SS 'tar'. However, it is not known what the r.^latiunship is. p. 81 - It is only correct to state that "Nonsmokers are exposed to ETS in indoor spaces" where smoking occurs. statement accordingly. Please modify p. 84 - The author states that "The utility cf Eq. 4 depends on the assumption of an air exchange rate" when in reality Eq. 4 depends on many more assumptions (no. of smokers, R:;P/cigt., sinks, etc.) The appropriateness and limitations of these assumptions are never clearly defined and they impact heavily on the usefulness of the proposed models. Gb (Z This reviewer fails to see how the example presented "illustrates -1 ~ the utility of models". Having data from one experiment which are N ~ 17, then fit to a model which has been adjusted Eor variables measured -1 11

in the experiment can in no way be used as validation of the general applicability of the model. Even i:n this case, when the actual air exchange is used, the model predicts 130% of the RSP actually determined. Combined with the observation that, on * average, RSP overestimates ETS by an additional 50% [T0;110],these facts severely hinder the general applicability of this model. The second example is equally flawed. The entire difference between daytime and evening RSP concentrations is erroneously attributed to ETS. The impact of persona]. activity other than smoking on the daytime RSP concentration i,s completely ignored. This so-called "pig-pen" effect (espoused pr:imarily by the authors of C hapter 6) should be explained as a further conforndcr of these data. The author "implies that ETS may diffuse throughout a large office building, exposing nonsmokers even in private offices" and offers * the data of Williams, 1985 [123]; as "support". The data of Williams et al., in fact, do not support t:ais implication. The results (Williams et al.) were miscalcu:lated and originally published incorrectly; corrections to the calculations were made * and subsequently published [34]. These corrected data were overlooked or ignored by the author and actually refute the observation. Assuming the analytical method of Williams et al. is sound, tneir data suppor` ey.:eedingly clean air in the office complex surveyed. p. 85 - The author states that from "li:mited field tests of * "[]" refers to Chaps. 5-8 Index. 12

the...model...it is clear that both models and observations based ,r1 on...monitoring...yield consistent results". This could not be further from the }~ ~t::- Simpl l stated, this; model has certainly not been validated and has yet to be used in even one incidence to provide reliable data in advance of actual air monitoring. p. 86 - What is the relevance of the narrative concerning the Mormons and the Seventh Day Adventists? p. 88 - The author correctly states that "In the •absence of...data...exposures can be estimated by models or by extrapolation from biological markers". The key word here is ~a "estimate" and it needs to be realized that the existing models have' yet to provide any indication of -re:producible, reliable estimates. In other words, the quality of these estimates is still ~ very much in question among scientists. This reviewer reiterates the comment provided to p. 79. ETS exposures can be assessed by air monitoring but only estimated by modeling, questionnaires, and the like. The statement "there are models in use...which can predict the concentrations of RSP from ETS to a reasonable degree of accuracy" is the opinion of the author and is not a generally held opinion. p. 89 - The author (inadvertently?) providea further evidence of the inappropriateness of calculated exposurEas- by stating "it has been calculated that a nonsmoker would inhale volatile nitrosamines 13

equivalent to...35 filter cigarettes." What evidence exists that this magnitude of exposure has, in fact, occu:cred? What the author alsc fails to include is the f?^_t that "no ep;c?emiological data exist linking human respiratory cancers to volatile nitrosamines." * ([g]; see also comment to Chapter 6, p. 71) The author makes further statements based on "the assumption that formation of cotinine...and clearance from the body does not differ substantially from smokers to nonsmokers". It is known, however, that formation of cotinine and clearance from the body do differ * substantially [2,45]. What is the relevance of the :.ote ddded regarding the RSP/nicotine ratio and the resulting calculated RSP? RSP/nicotine ratios are known to vary from 2:1 up to 100:1 across normal environments. This calculation serves no useful purpose. If you want to know RSP with any degree of confidence, you must measure it. This note should be deleted. p. 90 - The author presents a one-sided viewpoint ("may substantially underestimate") on the ability :)f nicotine absorption to predict exposure to other ETS constituents. It is at least equally likely (and never mentioned by this author) that nicotine absorption would overestimate exposure to other ETS constituents k,aGFc'. on possibilities of nicotine in the di.et: (see comment to p. 80) and of a detectable nicotine background in the absence of other ETS constituents (see comment to p. 58 [Ch. 5]). * "[]" refers to Chaps. 5-8 Index. 14

The author attempts to support his view of the appropriateness of RSP measurements for ETS exposure with a reference to the Surgeon General's report: "...*_:±e relationships of RUSP mea=»rPneTt.z to ETS are quite accurate". This comment is taken out of context and appears to be slanted solely for the author's purpose. The actual quotation from the Surgeon General's report is "At a practical level, the technology for measuring nicotine levels and RSP levels is available and accurate." The author should strive to remain objective in presentation of the relevant issues. p. 91 - It is not a consensus among researchers in the field that RSP is the best atmospheric marker of ETS exposure. This is the author's opinion. It is incorrectly stated that the cotinine measures reflect the actual dose of an ETS constituent. The confounding factors (diet, background nicotine) were never presented. The statement that cotinine measures "may substantially underestimate exposures to other constituents of ETS" is totally unfounded. As stated previously, it is equally likely (if not more so) that these measures will overestimate actual ETS exposures. It is presented as an advantage that RSP-based estimates are model- based_ In reality, this is its second most serious drawback (the first being that,RSP 41 ._ not specific to ETS). In discussing the GE) Qn drawbacks, the author fails to point out these limitations or the .~ .~ fact that RSP in nonsmokers is not absorbed in the same manner as N ~ 4D it is in smokers [2]. ~-A 15