Tobacco Documents Online

STEADY STATE CONCENTRATIONS OF COTININE ~ Renato L. Galeazzi Paul Daenens Matthias Gugger University of Berne, Department of Medicine, Iselspital, Berne/ Switzerland and University of Louvain, Department of Toxicology, Louvain / Belgium Address reprint requests to: R.L. Galeazzi, M.D. University of Berne Dept. of Medicine Inselspital 3000 B e r n e Switzerland 1

Nicotine is probably the most widely abused drug in the world. Its deleterious effects for public health are well known. There- forte the cigarette indust{y tries to manipulate the filters in such a way as to reduce the nicotine intake, without affecting the taste of the cigarette. These manipulations may not produce the wanted effects in the smoker, as a recent paper by Benowitz ~ et al (i) has shown. Therefore the real intake of nicotine into the body by the smoker shoed be adequately measurable to draw correct conclusions from epidemiologic studies, ormally ~z is ingested by inhalation of smoke from cigarettes. The deleterious effects on health depend on the amount of tar reaching the lung and the amount of nicotine reaching the circulation. Nicotine, at the average pH of cigarette smoke, is not absorbed in the upper respiratory tract but in the lung, where it is rapidly trans- ferred to the bloodstream (2). Because the tar-to-nicotine-ratio is quite constant for each brand of cigarette the determination of nicotine-intake could yield most of the necessary data for the epidemiology of smoking.. Nicotine, however, has a short half-life (3) in the range of Z~0 hrs, so that its concentration is rather sensitive to differences in sampling times.In addition it is very dependent on the nico- tine-yield of the last few cigarettes smoked. This is of great concern as individuals smoke with different intensities at diffe- rent times. Therefore, a marker should give an indication of average cummulative nicotine-intake over a suitable period of time.

Cotinine, the principale metabolite of nicotine (4) has a much longer half-life around 17 hrs .(5)0Its concentration in plasma has been used lately in several studies (i, 6-8) as a marker of n~cotine consumption. However, inferences from cotinine concen- trations as to the nicotine-intake can only be drawn if the relation between these t% variables is linear, i.e. if the meta- bolisme of nicotine to cotinine is not dose-dependent. In order~.~ to study this relation, we performed the following cotinine measurements in six healthy volunteers, which were infused seve- ral doses of nicotine i.v. over 4 days up to 480 ug/kg/day, we were able to show that cotinine concentrations in plasma were linearly related to nicotine intake (mean r2= 0.992+0.006 SEM).. Concentrations of cotinine showed little variation during and up to two hours after the last nicotine administration. Therefore, they can be used as epidemiologic markers of nicotine- intake if measured around the time of the last cigarette of the day. Material and Methods Six healthy volunteers, two females and four males were studied. Their mean age was 23.5 y (range 20 - 28), mean weight 61 kg (52 - 66). All gave their written informed consent after careful explanation of the study and the risks related with it. They were all smokers smoking at least 20 cigarettes daily for at least 12 months. They were not allowed to smoke during the time of the

study, but they continued to smoke (the same brand and the same number of cigarettes) inbetween the experiments. Before entering the study their health was checked by medical hi~tory, a physical examfnation, an EKG and a laboratory screen- ing including ESR, haemoglobin, leukocytes, thrombocytes, pro- thrombin time, blood urine nitrogine and creatinine, GOT, GPT, a urinary screening. They were hospitalized in the Dept. of Medicine at the Univer- sity Hospital (Inselspital) in Bern, Switzerland. Ordinary hospi- tal meals were served. Fluid intake had to be around 1 1/2 liters a day. Coffee and tea were freely allowed. The urinary pH was kept at 7 by giving sodium bicarbonate (NaHC03). Nicotine was administre~ntravenously as the free base in sterile sodium chloride 0.9% solution. It was provided by the Pharmaceutical Development Center of the College of Pharmacy of the Medical University of South Carolina, Charleston,S.C. (through T.D. Darby, Ph.D., USC school of Medecine, Columbia S.C.)in vials of 5 ml containing 1 mg/ml. Sterility and pyrogeni- city tests had been performed at the same place. Every 30 min the _amount of 5, i0, 15, 20 ug/kg bodyweight was given by continuous infusion over 10 min with a BraunR Melsungen precision pump. The first injection of the day was given at 7 a.m., the last at 7 p.m. The injections at each dose level were given for 4 days (Monday through Thursday). Every day the dose at 12.30 p.m. was skipped due to lunch. The doses per day were therefore 120, 240 360 and 480 ug/kg body weight. Before starting the first infusion in the morning and after

the last infusion of .the day 1 ml of the infusion liquid was saved for analysis of the nicotine con~ent. Blood samples of 5 ml were taken every day before the first of the daily doses. On the fmurth day (Thursday) blood samples were taken before, in the middle and at the end of the 7 p.m. infusion (e.g. during the last dose of nicotine of the last day), and then 5, 10, 20, 30, 45 min and i, 1.5, 2, 3, 4, 6, 10, 14, 16, 20, 24 hours after the end of the last dose. Blood was centrifuged immediately and the plasma was stored at -18°C until analysis. All the urine was collected, pH was measured in every sample ~to adjust the dosage of sodium bicarbonate)o~nd ztora~ at-~°C. Cotinine in plasma was determined with a newly developped GC-~5-- SIM procedure (P. Daenens et al.: in preparation). A deterium labelled analogue of cotinine (3-cotinine-methyl-CD3) was synthe- tized and used as the internal standard. 1 ml plasma iox~-m~=of-=l~4 di _ " _samp~e~Itx~ which a fixed amount (i00 ng) of internal standard was added were extrac- ted with methylene chloride in alcaline medium, the organic fraction dried over anhydrous sodium sulfate and carefully evapo- rated at 40~C. Prior to the injection, the residue was redissol- ved in i00 ~i of m~thanol and 1 ~i volumes injected onto the GC/MS system, operated in the SIM mode (m/z 176 for cotinine and m/z 179 for internal standard). A cpTM-wax 57 CB capillary column O of 50 meters was used /with spiteless injection.(Temperatures~' ~& injection port~ 260~C; column, 40~C for 1 min 40-150~C at ~ ' ~

35°C/min, 150-2~24°C at 15OC/min; DC/MS transfer lineh 225~C; ion source~ 150°C.") Carrier gas~elium 0.5 ml/min.). The coefficient of variation for a three day validation study o~ a pooled plasma sample" containing 85 ng of cotinine/ml varied between 3.6-5.4%. Nicotine in the injection fluid was determined by UV-spectropho- tofluorometry. The area under the plasma concentration vs time data of cotinine (AUC~) during the last dosing interval (~) were determined by the trapezoidal rule. The average concentration of cotinine at steady state (Css) was calculated using the following formula 6

Results The dose of nicotine injected, as measured in the infusion fluid differed little from the theoretical dose. The mean difference in percent of the theoretical dose was 0.283 % (range -5.3 to 12.7). The loss of nicotine in the injection fluid over the day, i.e. the difference between the concentration measured before the first application in the morning and after the last infusion at the end of the day was -0.59 % (range -2.5 to i). In Table 1 the mean steady state concentrations are given. As can be seen from F~. 1 a clear linear relation is given between the doses of nicotine injected over the day'for 4 days and the mean steady state concentration of cotinine during the last dosing interval. The intercepts of the regression lines were not statistically different from zero, therefore the regression line was forced through zero. The mean regression line was Cotinine conc (mcg/liter) = 0.783* Nicotine(mcg/kg/day) The smallest r2 from the individual regression lines was 0.988. The individual slopes ranged from 0.667 to 0.842. Figure 2 shows the time-course of the plasma-concentrations of cotinine after the different i.v.-doses of nicotine. There is a cleare plateau up to 2 hrs after the last injection.

Discussion Our results show, that the concentrations of cotinine are linear- ly related to the amount%icotine reaching the central circula- tion. This indicates linear kinetics for both nicotine and coti- nine up to the amounts of nicotine used and concentrations of cotinine measured. The regression line can even be used as a "standard-curve" in order to calculate the average nicotine- intake over the last four days from a single measurenent of cotinine-concentration in plasma. Although we used mean steady- state concentrations for our study, single determinations seem to be sufficient for epidemiologic use. The time course of cotinine during and after the last injection (or cigarette in real life) is flat for a period of about two hours. Therefore, single points at the end of "smoking-day" will yield reliable results. We did not continue to the highest possible doses of nicotine due to ethical reasons. Therefore, our results are only applicab- le to concentrations of cotinine of no more than 415 ug/l. How- ever, most of the cotinine levels found in epidemiological stu- dies (Gori) are in this range. In addition, as there is no hint at all as to a curvature in the relationship nicotine to cotihine, extrapolation for little higher concentrations may be performed, although with less reliability. Strictly speaking, the regression line presented here (eg. i) can only be used in a population similar to the one from which our subjects were drawn, because to give reliable results clear- ances of both nicotine and cotinine ha~e to be the same. However,

if comparative experiments are made it will yield at least rela- tive results. An example of its application to epidemiologic data can be shown using data published by Gori and Lynch (6). - Theymeasured cotinine ievels in a large group of smokers, which were given three different brands of "low-tar" cigarettes. Mean cotinine levels (measured in the late afternoon), mean number of cigarettes, and the FTC-rating, as given in their .... article, are shown in table II. From these data, using equation i, we have calculated a mean daily nicotine intake (assuming a mean body weight of 65 kg). Dividing this value by the mean number of cigarettes smoked per day, one arrives at a mean nico- tine content per cigarette. As can be seen in table II, this value differs by almost an ~t order of ~agnitude from the official FTC-rating. However, as the ratio between the cigarettes approaches the FTC-rating, it can probably be regarded as an appropriate relative rating, at least in the "low tar" group of cigarettes We think that cotinine measurements and hence determination of nicotine-intake should be of great help in epidemiological stu- dies of nicotine consumption. 9

Legend to Fi@ures Figure 1 S~eady state concentratiohs of cotinine vs the nicotine dose. The nicotine doses indicated were given over 12 hrs for four days. Thecotinine levels were measured during the last dosing interval on the fourth day. Shown are the mean values (+/- SD), together~ _ with the mean regression line. Fi@ure 2 Time course of the concentrations of cotinine during and after the last i.v.-nicotine injection. The nicotine doses indicated were given over 12 hours for four consecutive days. i0

References i. Benowitz NL, Hall SM, Herning RI, Jacob III P, Jones RT, - OsmanAL ~ Smokers of low-yield cigarettes do not consume less nicotine. N Engl J Med 1983b; 309:139-142 2. Armitage AK Some recent abservations relating to the absorption of nicotine from tobacco smoke. In: Dunn WL, ed. Smoking behavior: Motives and incentives. Washington DC: Winston, 1973:83-91 3. Rosenberg J, Benowitz NL, Jacob P, Wilson KM Disposition kinetics and effects of intravenous nicotine. Clin Pharmacol Therap 1980; 28:517-522 4. Beckett AH, Gorrod JW, Jenner P The effect of smoking on nicotine metabolism in vivo in man. J Pharm Pharmacol 1971; 23 (Suppl.): 62S-67S 5. Benowitz NL, Knyt F, Jacob III P, Jones RT, Osman AL Cotinine disposition and effects Clin Pharmacol Ther 1983; 34:604-611 ii

6. Gori GB, Lyich CL Smokers intake from cigarettes in the l-mg Federal Trade Commission tar class. - Regulatory Toxicol Pharmacol 1983; 110-120 7. Gritz ER, Baer-Weiss V, Benowitz NL, Van Vunakis H, Jarvik ME Plasma nicotine and cotinine concentrations in habitual smoke- less tobacco users. Clin Pharmacol Ther 1981; 30:201-209 8. Matsukura S, Sakamoto N, Seino Y, Tamada T, Matsuyama H, Muranaka H Cotinine excretion and daily cigarette smoking in habituated smokers. Clin Pharmacol Ther 1979; 25:555-561 12

Table 1 Mean steady state concentration of cotinine in plasma during the last dosing interval on the fourth day (ug/liter) Dose of nicotine in ug/kg/day (in ug/kg per injection) No 1 No 2 No 3 No 4 No' 5 No 6 120 (5) 79.15 118.08 109.33 85.08 71.08 89.50 240 (10) 168.33 204.75 192.33 195.75 167.42 182.00 360 (15) 240.92 276.41 292.33 314.83 249.42 291.00 480 (20) 407.25 413.16 415.00 376.58 311.67 360.75 X (SD) 92.04 (18.1) 185.1 (15.2) 277.49 (28) 380.74 (40.29) 90 UO [O

Table II ~ Mean nicotine content of three different brand of cigarettes using data colle~ted by Gori and ~Lynch (6) and our/~e~ I. :otinine level ng/ml Cigarettes per day • FTC rating Mean nicotine per c~garette mg Ratio to C Brand A B C 301 204 29.0 30.5 0.18 0.861 0.Ii 0.55 1.57 1.01 208 31.5 0.i0 0.548 1.0

÷ t I I ! I © o~.~I'/S°~ 03 ~o ouo3

Table II Mean nicotine content of three different brand of cigarettes using data colle ted by Gori and Lynch (6) and our ~ i. Brand Cotinine level ng/ml Cigarettes per day A B C 301 204 208 29.0 30.5 31.5 FTC rating .. 0.18 0.ii 0.I0 Mean nicotine per cigarette mg 0.861 0.55 0.548 Ratio to C 1.57 1.01 1.0