Tobacco Documents Online

Topical Report NCI/S&HP/ORNL #67 SITE VISIT I TO THE HAZLETON LABORATORIES COFACTOR INHALTION BIOASSAY Reston, Virginia, February 23, 1978 3-16-7£3 R. A. Jenkins and B. E. Gill Tobacco Smoke Research Program Bio/Organic Analysis Section Analytical Chemistry Division Oak Ridge National Laboratory Oak Ridge, Tennessee 37830 Interagency Agreement (ERDA-PdIFf/NCI) No. 40-485-74, Part II Internal (aRNL) Contract Charge No. 3390-0224 Intended for informal communication with project management only. Confidential until published or released by author.

SITE VISIT I TO THE HAZLETON LABORATORIES COFACTOR INHALATION BIOASSAY , R: A. Jenkins and B. E. Gi1l Summary. A first visit was made to the new tobacco smoke-cofactor inhal- ation bioassay at Hazleton Laboratories to assess the relative importance of the contribution of several factors to any potential variation in carbon monoxide dose in the animal exposure groups. The data obtained on that visit indicate that the added gas delivery systems of the smoke exposure machines perform very reproducibly and cannot be a significant source of variation in carbon monoxide dose. The most important contribution to the small variation that is noted is the number of puffs required to consume a given cigarette to a 23-mm butt length. This value will undergo normal cigarette-to-cigarette variation, and without expensive weight and resistance- to-draw selection of cigarettes, cannot be tightly controlled. In order to reduce the small difference in CO dose between the two cofactor addition ex- posure groups, it was recommended that the per-puff added gas delivery of the Code 23 exposure systems be slightly reduced. Introduction. At the request of NCI Smoking and Health Program management, ORNL staff visited the tobacco smoke-cofactor inhalation bioassay being con- ducted at Hazleton Laboratories (HL) on February 23, 1978. The primary ob- jective of this visit was to assess the magnitude of the cofactor (carbon monoxide) dose which the animal groups would be offered, and determine the extent and source of any variability in that dose. Experimental. This is a new inhalation study. It involves the chronic ex- posure of male beagle dogs to whole cigarette smoke and smoke enriched with carbon monoxide (CO). For the cofactor studies, the smoke exposure system being employed is the ADL/II-cuffed cannula system modified with an external gas delivery system. The ADL/II has been affixed with a valve whose operation has been synchronized to the travel of the smoking machine's sliding cylinder. The cofactor gas (,ti20'% CO in nitrogen) is metered through an adjustable needle valve, and into a needle which penetrates a small piece of rubber tubing be- tween , . tween the outlet of the ADL/II and a glass bulb attached to the head of the system stand tube. Constancy of the flow of added gas is ensured by (1) i

constant pressure at the second stage of a two-stage regulator on the gas supply tank and (2) constancy of restriction of the met.ering needle valve. As the sliding cylinder reverses its_direction, the metering valve shuts off, and the gas aliquot enters the bulb at the head of the stand tube fractions of a second before the smoke enters. The animal then inhales the smoke-gas bolus as he would normal cigarette smoke. Most experimental details and procedures have been described previously, and can be found in reports covering earlier visits to HL (Topical Reports NCI/S&HP/ORNL #24, 29, 33, 36A, 46, 64). A soap bubble displacement meter was used to measure the volume of added cofactor gas. The needle was removed from the stand tube of the ADL/II and inserted into a tube connected to the meter. The volume displaced by the bolus was noted and recorded. Results and Discussion. The dose of carbon monoxide, per cigarette, which the animals exposed to smoke plus CO are offered is a function of four dis- tinct factors: 1. Concentration of carbon monoxide in the added gas. This is constant, assuming the integrity of the supply tank is main- tained. 2. The volume, per puff, of the added gas, which may or may not vary among individual exposure systems. 3. The length of time of exposure to the added gas. This is directly proportional to the amount of time the machine is operating, which, in turn, is a function of the number of puffs which the cigarette is smoked. 4. The amount of carbon monoxide in the smoke of the cigarette itself. This is subject to normal cigarette-to-cigarette variability. An attempt was made on this site visit to assess the variability of these factors. The results of the assessment are discussed below. The volumes of gas added, per puff, for selected machines are tabulated and compared on Table I. Clearly, the amount of added gas delivered to the ADL/II stand tube is highly reproducible* from machine to machine. The relative standard deviations for each +code are less than 4%. The absolute magnitude of the de1 i very i s about 20 ml per puff, about 7 3% hi gher than the targeted

addition of 17.5 ml per puff. The difference is probably due to the use of a syringe barrel displacement procedure at Hazleton to determine gas addition volume. There may be a resistance to displacement of the syringe barrel which would increase the pressure of the gas in the syringe. The 17.5 ml displaced volume measured with the syringe at slightly elevated pressure might be equivalent to a volume of 20 ml at ambient pressure. In summary, while the amount of cofactor gas may be slightly greater than that targeted, the machine-to-machine variation in this gas volume is so small that it cannot be responsible for offered dose variation or for differences in the carboxyhemoglobin levels among animals in a given exposure group. The TPM and nicotine deliveries of cigarettes smoked on ADLJII's at Hazleton are compared on Table II with those of cigarettes from the same batch returned to ORNL and smoked under analytical conditions. Interest- ingly, the Code 23 variant produced more nicotine but less TPM at the HL than under analytical conditions. The carbon monoxide deliveries under analytical conditions are very similar to those determined previously under the same conditions (see letter to E4. B. Coate from R. A. Jenkins, July 26, 1977, attached), suggesting that no gross differences exist between the original control cigarettes and those sampled from the bioassay. The amounts of CO, nicotine, TPM and C02 delivered to the cannula under both normal smoking and cofactor addition protocols are presented on Tables III and IV for the Code 13 and 23 exposure groups, respectively. In addi- tion, amounts of CO presumed to have been added for each machine-cigarette combination are also presented. Several important points should be noted. First, the amounts of TPM, nicotine, and C02 delivered to the cannula for a given cigarette variant are essentially the same under both normal smoking and cofactor addition conditions. This suggests that both types of machines (cofactor and normal) are smoking cigarettes in a similar manner. Secondly, the amounts of carbon monoxide delivered to the cannula under normal smoking conditions are the same for both codes. This similarity has been noted previously (see text of attached letter) for CO delivery on the ADLJII, but the magnitude of the delivery was signifiecantly different. Presumably, both types of cigarettes generate'COrwhich is delivered in equal amounts to the cannula under cofactor addition conditions also.

The final point is the most important. From the data on Tables III and IV, it is clear that there is quite a bit of variation in the amount of CO calculated to have been added for each machine-cigarette combination. This calculated value is a product of (a) the concentration of carbon monoxide in the cofactor gas (20.3%), (b) the volume of gas added per puff for each in- dividual machine (see Table I), and (c) the number of puffs taken to consume each cigarette. Since (a) and (b) are virtually constant, clearly the greatest contribution to variation in the amount of added CO for each individual machine-cigarette combination is the number of puffs required to smoke the cigarette. As is evident from Tables III and IV, this puff number can vary greatly even within a given variant code. In addition, the amount of added carbon monoxide appears to be greater for the Code 23 group than for the Code 13 group. This is due to the differences between the two variants in the number of puffs required to smoke the cigarette to a 23-mm butt length. A summary of the per-cigarette offered doses for two of the more important smoke constituents is made on Table V. Ideally, the amount of carbon monoxide offered the animals receiving the added gas dose should be the sum of the ciga- rette CO contribution and the amount of CO calculated to have been delivered to the exposure system. For the Code 23 group, the sum of the average deliveries 16.0 + 59.4 = 75.4 ml CO/cigarette) compares well with that which was observed at the cannula (76.7 ml/cigarette). For the Code 13 group, the anticipated dose (66.7 ml,Ocigarette) also compares well with the observed dose (67.2 ml/cigarette). The ratios of the carbon monoxide dose for the cofactor addition group to that of the normal smoking group are 4.4:1 for the Code 13 variant and 4.8:1 for the Code 23 variant. Both of these ratios are substantially greater than the targeted value of 3:1, primarily for two reasons. First, the added gas delivery measuring system in use at I-fL permits a slightly greater amount of CO to be added for each puff. Secondly, both variant codes do not deliver as much carbon monoxide to the cannula as was anticipated from analytical smoking or exposure system sm®king at ORNL (see attached letter). Again, the primary reason for the CO dose differences between the two cofactor addi- tion groups is the difference'in the puff numbers of the two variants. Recommendation. The preceding information was reviewed by phone with Dr. W. Coate of Hazleton and Dr. T. Nightingale of Enviro Control, Inc. Based on those discussions, the following recommendations were made:

In an attempt to overcome differences in CO dose for the two cofactor addition groups, the per puff delivery of carbon monoxi•de should be slightly reduced for machines used to smoke the Code 23 cigarettes, such that the total added CO delivery over the greater number of puffs of the Code 23 var- iant will be equal to that of the machines smoking the Code 13 cigarettes in a fewer number of puffs. No other changes in the smoking or exposure protocol should be made. In addition, a more accurate measurement system for added gas delivery volumes, such as a soap bubble meter, should be used by Hazleton.

TABLE I Per-Puff Machine°Gas Addition Volumes* Cofactor Addition Machines Cigar Co ette de Machine No. Added Gas Vo1ume, ml/puff 13 34 19.6 28 19.9 40 19.4 14 20.3 12 20.3 39 20.0 Average 19.9 ± 0.4 23 25 19.9 I 5 20.0 10 19.4 9 18.5 1 20.6 19 19.5 Average 19.7 ± 0.7 *"Gas" added is a portion of a standard which is 20.3% CO in nitrogen.

TABLE II Comparison of Cigarette Deliveries: Input to ADL/II at HL vs Analytical Smoking Machine at ORNLa Site Visit C-I Cigarette TPM Nicotine C0, C02 Machi ne Code # Puffs mg/ci g mg/ci ~ m1 /ci g ml Jci g ADL/II 13 12.7 t 0.8 38.2 ± 6.4 0.19 ± 0.03 b b Analytical 13 11.5 ± 0.5 43.7 ± 0.5 0.23 ± 0.02 24.4 ± 0.7 47.9 ± 2.2 ADL/II 23 14.8 ± 1.5 44.7 ± 6.0 2.62 ± 0.18 b b , Analytical 23 16.4 ± 0.3 53.3 ± 0.8 2.20 ± 0.04 30.6 ± 0.6 65.6 ± 1.1 I aCigarettes smoked on Phipps and Bird analytical smoking machine, at 2 puffs per minute, eight cigarettes per code. bInput deliveries of CO and C02 for cigarettes smoked on ADL/II are not conveniently obtainable. However, since little loss of permanent gases occur within machine, outputs of CO/CO2 for each cigarette code (from Tables III and IV) can be compared with the analytical values on this table. r084£&68

TABLE III Cannula Exit Deliveries ADL/II Systems Smoking Code 13 Cigarettes Machine No. # Puffs TPM, mg/cig Nicotine, mg/ciq C0, ml/cig Calculated C02 Amount of ml/cig Added C0, ml ~s 16 14 23.9 0.15 15 35 /" 17 15 23.6 0.14 14 37 Normal / 24 13 25.9 0.12 15 36 Smoking 32 14 28.2 0.18 20 41 3 11 23.8 0.16 13 29 'Average 13.4 ± 1.5 25.1 ± 2.0 0.15 ± 0.03 15.4 ± 2.7 35.6 ± 4.3 34 16 25.0 0.17 74 35 63.7 28' 13 25.0 0.22 68 31 52.5 Smoking with 40` I 12 20.5 0.14 62 31 47.3 Added Carbon 14 11 23.4 0.14 62 33 45.3 Monoxide 12 11 22.1 0.14 69 32 45.3 39 13 23.1 0.17 68 30 52.8 Average 12.7 t 1.9 23.2 ± 1.7 0.16 ± 0.03 67.2 ± 4.6 32.0 ± 1.8 51.2 ± 7.0 i Z ~'q4E&fiB

TABLE IV Cannula Exit Deliveries ADL/II Systems Smoking Code 23 Cigarettes TPM, Nicotine, C0, C02 Calculated Amount of Machine No. # Puffs mq/cig mg/cig ml/cig ml/ciq Added C0, ml 26 15 31.2 1.95 16 42 38 16 33.5 1.99 17 44 Normal 30 16 35.6 2.11 20 47 Smoking 36 13 24.6 1.49 12 36 37 13 27.1 1.59 15 36 6 14 28.1 1.53 16 39 Average 14.5-F 1.4 30.0 ± 4.2 1.78 ± 0.27 16.0 ± 2.6 40.7 ± 4.5 2 5 15 31.7 1.90 81 44 60.6 5~ 15 28.2 1.53 76 37 60.9 Smoking with 10 16 29.6 1.88 80 36 63.0 Added Carbon 9 12 19.1 1.69 64 26 45.1 Monoxide 1 17 37.1 1.47 85 35 71.1 19 14 29.5 1.76 74 37 55.4 Average 14.8 ± 1.7 29.2 ± 5.9 1.71 ± 0.18 76.7 ± 7.3 35.8 ± 5.8 59.4 ± 8.6 EVs4=9