Smoking and Tobacco Control Monograph No. 7 
volume inhaled, significantly affect the smoker's exposure to tobacco-specific 
nitrosamines. According to the investigators, 
The medium-tar cigarette using standard smoking conditions 
delivered TSNA values that were close to the calculated average 
intake by smokers. The calculated average TSNA intake for the 
low-tar cigarette, however, was about double the value determined 
under standard smoking conditions (Fischer et al., 1989, p. 1065). 
The researchers concluded that 
since the standard smoking conditions cannot reflect the real 
behavior for low- and very-low-tar cigarettes, especially with 
respect to the total inhalation volume, risk evaluation has to 
consider the increase in TSNA intake with increasing total 
volume (Fischer et al., 1989, p. 1065). 
In a subsequent study, Fischer and colleagues (1991) investigated 
170 types of American, European, and Russian cigarettes. The findings 
revealed that the amounts of two TSNAs — NNN (N-nitrosonornicotine) 
and NNK (4-methylnitrosamino-l-[3-pyridinyl]-l-butanone) — in cigarette 
smoke are not correlated with tar or nicotine delivery and the amounts of 
TSNAs in mainstream smoke are related to the amount of preformed 
nitrosamine in the tobacco. 
In an investigation of compensation behaviors among smokers switching 
to lower delivery cigarettes, Robinson and coworkers (1983) noted 
disproportionate increases in HCN levels. The researchers concluded that 
machine-determined "standardized" deliveries do not reflect potential 
exposure to HCN. 
Rickert and Robinson (1981), in a study of delivery by low-hazard and 
high-hazard brands and actual levels, found that differences in HCN and 
CO yields of the two different delivery types varied much more widely than 
actual levels of COHb and plasma thiocyanate obtained from smokers of 
each. High-hazard cigarette smokers had nearly four times the HCN of 
low-hazard cigarette smokers; however, the actual levels differed by only 
20 percent. These differences were not statistically significant, possibly 
due to small sample size (n = 31). 
Rickert and colleagues (1983) looked at variations in smoking patterns 
and reported that HCN delivery is influenced by blocking of ventilation 
holes and, to a lesser degree, by puff duration, puff volume, and butt length. 
Blocking half the ventilation holes increased HCN yield by 70 percent; 
covering all the holes produced a 250-percent increase in yield. Fhese 
investigators determined that HCN yield for the cigarette brand investigated 
ranged from 5 to 241 pig, depending on variations in smoking parameters, 
although the mean HCN yield was 39 ng. For 115 Canadian cigarettes, the 
average HCN yield varied from 2 to 233 pig. Fhis impact of smoking pattern 
on HCN yield was cited by Rickert and colleagues as a possible explanation 
272 
