Comparison of Select Analytes from Different Laboratories in Tobacco and Smoke for Commercial Cigars Across a Range of Designs

Analytical research relevant to cigar products is relatively limited compared to other tobacco product categories and very few standardized methods are in place. In recent years, scientific and regulatory interests in cigar testing have increased. Thus, there is a need for increased foundational knowledge in the product space. The objective of this work was to characterize cigars across a range of design features to understand relative magnitude of results and to evaluate analytical variability among and between three laboratories using their own practices and methodology.

Commercial cigars across a broad range of design characteristics were evaluated for tobacco and smoke harmful or potentially harmful constituents (HPHCs) typically applied to cigarettes and smokeless tobacco products (i.e., FDA-Center for Tobacco Product’s abbreviated list). Design features such as size, manufacturing technique (machine vs. hand), and tipping (untipped, filtered, and mouthpiece designs) were included in the study matrix. For smoke analytes, a currently accepted regime for cigar smoking was employed. Cigars from the same lots were tested by three independent laboratories using their own methodologies to assess lab-to-lab differences across 5 physical parameter measures, 10 tobacco analytes, and 21 smoke analytes. Thus, the overall study was a randomized design with a two-factor arrangement: cigar type (with six levels) and laboratory (with three levels). Calculation of Lab Product Range (LPR) allowed for a comparison of magnitude in results across the product set. Data analysis included determination of relative variability (% RSD, Relative Standard Deviation) for replicate testing among the products. Lab Range (LR) was calculated to compare spread in results among the laboratories for each product. Low LR and similar LPR among the labs is an indicator of laboratory consistency.

LPR results ranged from 45% for length to 445% for smoke ammonia. For example, tobacco weight, at approximately 260% LPR, ranged from 1 g/cigar to 18 g/cigar. Tobacco NNN, LPR 328%, ranged from 2000 ng/g to 27,000 ng/g. Replicate variability (% RSD) of physical characteristics, tobacco, and smoke analytes was much higher for the cigar products than previously reported for typical cigarette products. For example, tobacco weight % RSD was as high as 11%. Commercial factory-made cigarettes are typically reported at < 1% RSD.

Lab Range was as low as < 1% for cigar length and as high as 247% for tobacco NNK. Smoke nicotine, one of the few analytes tested using standard methodology and subject to routine collaborative studies, had an LR as high as 73%. Tobacco BaP and carbonyls LR results were very inconsistent between labs. For example, acetaldehyde for Sample D was reported as 0.29, 0.81, and < 0.10 µg/g for Lab 1, Lab 2, and Lab 3, respectively.

Cigars with a broad range of design features were shown to have relatively high variability, a wide magnitude in analyte levels and smoking characteristics, and in some cases quite distinct results from lab-to-lab. Some analytes, such as BaP and crotonaldehyde, were determined to be inadequate for characterization in cigar tobacco due to a lack of quantifiable results. Based on LR results, smoke ammonia and carbonyls were found to lack robust methodology.

The results of this study support the need for increased standardization for smoking analyte methods and the use of reference products across studies to improve understanding of product differences and the contribution of analytical variability. Additionally, in the absence of foundational data, the results are cautionary regarding using isolated data sets for characterization via HPHC testing for cigars, especially hand rolled products.