This study included 49 consecutive healthy participants (34 males and 15 females) who volunteered at the Department of Otorhinolaryngology-Head and Neck surgery, Konkuk University Medical Center, from April 2015 to May 2016. We excluded participants with symptoms or signs suggesting other sinonasal diseases or disorders of the CNS. Thus, participants with obstructed olfactory clefts or other diseases entities that might cause conductional olfactory dysfunction such as nasal septal deviation or nasal polyps were excluded. Those with a history of posttraumatic and post-infectious olfactory dysfunction or other miscellaneous types of olfactory disorder with decreased sense of smell were also excluded. All participants were fully notified of the contents of this study. They agreed to participate after obtaining informed consent under approval of Institutional Review Board (IRB No. KUH1110063).

This study was a double-blinded clinical trial. Subjects underwent a series of test for 2 days. Considering the half-life of caffeine which is known to be 4–5 hours with modest intake [21], each subject was restrained from caffeine consumption at an extended period of 10 hours for sufficient clearance. On the first day, individuals were tested for their olfactory visual analog scale (VAS) score, minimum cross-sectional area (MCA) measured by acoustic rhinometry, KVSS II and d2 test prior to coffee consumption. These same tests were then conducted 30 minutes after drinking a cup of coffee containing 72 mg of caffeine or decaffeinated coffee with 8.4 mg caffeine (KANU, dark roast Americano or KANU, decaffeine Americano; Dong Suh Food Co., Seoul, Korea). On the second day, the same experiment was done. However, each individual was asked to drink the opposite type of coffee as they did the day before. The type of coffee initially administered was randomly selected. Results of the olfactory function and d2 test before and after caffeinated and decaffeinated coffee intake were analyzed respectively to determine statistical significance. Because neuropsychological tests are affected by circadian rhythm, all subjects were tested at 17:00 PM consistently.

A survey was conducted to evaluate subjects using VAS. Results of acoustic rhinometry MCA was measured. All participants were initially examined by an otorhinolaryngologist (KJK) with a 2.7 mm×0° nasal endoscope to check the patency of the olfactory cleft, the existence of nasal polyps or nasal obstruction due to septal deviation, mucoid or purulent discharge, and any other signs of sinonasal diseases. Olfactory function was evaluated with KVSS II that offers a comprehensive measurement of individual’s olfactory function. It consists of three different sets that measures T (olfactory threshold), D (discrimination), and I (identification). These tests are performed in the same way as the Sniffin’ Stick test. Threshold was defined as the concentration where n-butanol (highest concentration 4%, 1:2 serial dilutions to 16 steps) was correctly identified four times in a row from two blanks. For discrimination, triplets of odorants (two of them are identical, one is different) are presented and subjects have to choose the different one. In the identification test, subjects were asked to choose from 16 items that was recorded with as score ranging from 0 to 16. Theses scores were subsequently added up as the “KVSS II total score (TDI score).” Total scores of 0–20, 20–27, and 28–48 were defined as “anosmia,” “hyposmia,” and “normosmia,” respectively [22,23].

Evaluation of participant’s attention was measured with the d2 test, developed by Brickenkamp and Zillmer [24]. The d2 test measures attention with a time-limit. By differentiating similar optical stimulus, values such as processing-speed, regulation obedience, and quality of performance can be measured. It is a tool that evaluates an individual’s attention and concentration ability. The d2 test is composed of 658 stimulants, each of which is categorized into 16 different subtypes. For instance, alphabet “d” or “p” are marked with one to four dashes and subject is required to select and respond to only d’s with two dashes (d”). The d2 test is composed of several different parameters. For instance, the total number of items processed measures processing speed, while the total number-total error (TN-E), which is the number of correctly marked characters minus the number of incorrectly marked, evaluates overall performance. Concentration performance, which is measured by total number of correctly cancelled minus total number, also represents overall performance. Fluctuation rate is a measure of consistency and stability of the performer.

Statistical analyses were performed using SPSS ver. 17.0 (SPSS Inc., Chicago, IL, USA). To determine differences in VAS score, acoustic rhinometry, KVSS II, and d2 test between results of before and after caffeinated and decaffeinated coffee intake, continuous variables were tested with paired t-test. A 2-tailed P-value of less than 0.05 was considered statistically significant.

This study was carried on 49 patients (34 males and 15 females) with a mean age of 27.7±2.8 years (range, 24 to 35 years). The average subjective VAS score of participants was 7.9±1.2. Results of the acoustic rhinometry MCA without any administration of caffeine or decaffeinated drink showed an average of 0.6±0.3 cm² on the right side of the nasal cavity and 0.6±0.3 cm² on the left side. The mean KVSS II total score was 29.0±4.8. Average TDI score was threshold, 6.3±2.5; discrimination, 10.5±2.3; and identification, 12.1±2.1. In d2 test, the following results were obtained: mean score of total number of items processed, 608.8± 57.3; omission error, 17.6±22.5; commission error, 1.2±3.3; total number of errors, 17.8±22.7; TN-E, 590.90±68.64; concentration performances, 249.3±40.7; fluctuation rate, 7.6±6.9 (Table 1).

Although the VAS score and nasal cavity area showed slight increase after caffeine intake, there was no statistically significant difference in VAS scores between before and after caffeinated coffee intake or between before and after administration of decaffeinated coffee (P=0.19, paired t-test). Acoustic rhinometry showed no significant difference in MCAs between right and left sides (P=0.35) (Table 2, Fig. 1). Subjects showed no significant improvement of olfactory function following caffeine administration. Results did not show any significant difference in KVSS II total score, odor threshold, odor discrimination, or odor identification between before and after caffeinated or decaffeinated coffee intake (KVSS II total score, P=0.19; threshold score, P=0.79; discrimination score, P=0.35; identification score, P=0.69) (Table 3, Fig. 2).

Results of d2 test showed statistically significant differences in the parameters of total number of errors and omission errors (total error, P=0.01; omission error, P=0.01). After caffeine intake, the amount of omission error was decreased to –10.0±13.1 than that before intake. After decaffeinated coffee intake, omission error decrease was –4.2±9.6 (P=0.01). The total number of errors after caffeine intake was decreased to –9.8±13.2 compared to that before intake. After decaffeinated coffee intake, the total number of errors decrease was –3.9±10.0 (P= 0.01). There was no significant difference in the remainder parameters of d2 test (total number of items processed, P=0.82; commission error, P=0.10; TN-E, P=0.24; concentration performances, P=0.89; and fluctuation rate, P=0.49) (Table 4, Fig. 3).