The ZIA factors for lozenges from the following clinical trials were evaluated, and the results of each evaluation follow. Use of the ZIA equation and ZIA factor data provided a standard means (ZIA values) of evaluating lozenges of different designs and manufacture. After a standard means of evaluating lozenges was developed, evidence of dose-response linearity became sufficiently evident to warrant rejection of the null hypothesis that there is no relationship between ZIA values and reductions in duration of common colds.
In the original double-blind, placebo-controlled, clinical trial by Eby et al. at the University of Texas at Austin, 23-mg zinc (175-mg zinc gluconate) lozenges shortened natural colds by an average of 7 days compared to placebo.(1) Patient reports in the published 1984 report were restricted by the editors to 65 patients who had been sick for 3 days or less prior to enrolling in the trial. The half-life of common colds in the zinc-treated group was 2.7 days compared to 7.5 days for colds in the placebo-treated group, resulting in a difference of 4.8 days in half-lives. Calculations using half-lives to determine average durations of exponentially decaying colds (half-life / ln 2) yielded an average duration of 3.9 days for zinc-treated colds and 10.8 days for placebo-treated colds, resulting in a reduction in average duration of 7 days. Statistical significance for the reduction in average duration of colds was high (p = 0.10 at 12 hours, p = 0.008 at 24 hours, and p = 0.0005 at day 7).(1) Lozenges were astringent, slow dissolving, unflavored tablets having no soluble ingredients other than zinc gluconate. Common cold symptom severities were similarly reduced. Lozenges tasted chalky, dry, and bland, but they were not bitter. Lozenges had a similar aftertaste. These zinc gluconate lozenges had a ZIA value of 100. Results of the entire 1984 experimental group of 146 patients have been published.(25)
The effect of pleasant-tasting, 23 mg zinc (175 mg zinc gluconate), fructose-based, hydroxypropyl methycellulose-bound, wet-granulated, compressed lozenges were reported by Al-Nakib et al. at the MRC Common Cold Unit in Salisbury, England, in 2 double-blind, placebo-controlled, clinical trials (a prophylactic and a therapeutic trial) in 1987. All patients were quarantined for 2 days prior to the trial and during the trial. Tests demonstrated that the placebo and zinc lozenges were indistinguishable in taste and appearance.(2)
Pretreatment with lozenges had a prophylactic effect on occurrence of rhinovirus-2 colds in 57 patients. Zinc medication reduced the mean daily clinical score on successive days by about one third when compared to placebo, and was statistically significant (p = 0.05) on the second day after virus challenge. The mean clinical score was reduced from 8.2 in the placebo group to 5.7 in the zinc group. Zinc treatment compared with placebo treatment resulted in lower mean weight of nasal secretion and fewer patients who excreted virus.(2) The study terminated prior to determining the therapeutic effect of the lozenges, consequently a therapeutic trial was conducted.
The lozenges shortened clinically induced human rhinovirus-2 colds in the therapeutic trial involving only 12 subjects by an average of 4.8 days compared with placebo, with statistical significance evident during the 6 day study. The mean daily clinical scores of zinc-treated patients were clearly reduced compared with placebo on days 3 - 6 (p < 0.01 - 0.05). The total mean clinical score for the 6-day study was 27.2 in the zinc-treated group, and 41.0 in the placebo-treated group. Zinc lozenge treatment substantially reduced mean weight of daily nasal secretions compared with placebo (P < 0.05) throughout the study. The total mean number of tissue papers used by the zinc- and placebo-treated groups was 14.3 and 21.7 respectively (P < 0.01). Mean tissue use on day 4 of zinc versus placebo treatment was 1.42 tissues versus 2.75 (P < 0.05), respectively. Mean tissue use on the fifth day of zinc versus placebo treatment was 0.25 tissues versus 2.17 (p < 0.01), respectively. Mean tissue use on the sixth day of zinc versus placebo treatment was 0.33 tissues versus 1.67 (p < 0.05), respectively.
By day 4, common cold nasal secretions in the zinc-treated group were comparable with preinfection levels, while nasal secretions in the placebo-treated group remained high from the fourth through the six day of the trial -- identical to nasal secretion scores of the zinc-treated patients on day 3. These zinc gluconate lozenges had a ZIA value of 43.9. Although Zn2+ ion release was the same as in the original 1984 trial, the ZIA value was lower because lozenges dissolved faster, produced viscous saliva, and produced more saliva than the original 1984 lozenges, thereby resulting in a lower concentration of salivary Zn2+ ions and some occlusive effects. These astringent lozenges were sweet and highly flavored and were well-tolerated by patients.(2) Fructose is the only sweet tablet base that does not increase bitterness of zinc gluconate upon lozenge aging.(25)
In contrast, the effect of bitter, low-dosage, 11.5 mg zinc (87.5 mg zinc gluconate) sucrose-, fructose-, mannitol-, sorbitol-based lozenges were reported by Smith et al. in 1989 in a double-blind, placebo-controlled, clinical trial of natural colds at multiple centers in the U.S. in 174 patients. The number of zinc-treated ill patients was reduced on day 7 by 12.6 percent compared to placebo (p = 0.09), and the severity of zinc-treated colds on days 5 - 7 was reduced by 7 - 8 percent (p = 0.02).(31) These compressed zinc lozenges reduced the average duration of common colds by 1.6 days compared with placebo. The ZIA value of these mildly astringent lozenges was 25 because of low zinc content, rapid lozenge dissolution, and bitterness-induced deviation from the clinical trial protocol (one lozenge each 2 hours rather than two lozenges each 2 hours -- personal communication, C. B. Goswick, M.D., Student Health Center, Texas A&M University, 1989). Bitterness resulted from a slow, solid-state reaction between zinc gluconate and non-fructose sweeteners. Greatly confounding the analysis were at least 16 ingredients present in the lozenges according to the zinc lozenge manufacturer (personal communication, B Bannen, Product Development Specialist, General Nutrition Products, Greenville, SC, June 27, 1989).
In 1990, Weismann et al. likewise reported no efficacy from low-dose zinc gluconate lozenges in a double-blind, placebo-controlled, clinical trial of natural colds in Denmark involving 130 patients. These non-astringent, maltitol-based, hard-boiled candy lozenges contained 4.5 mg zinc (31.3 mg zinc gluconate), and had a ZIA value of 13.4. Low zinc dosage was used to avoid the extreme bitterness of zinc gluconate in maltitol-based lozenges.(32)
Non-astringent zinc orotate (37 mg zinc) compressed lozenges used in conjunction with 10 mmol zinc gluconate nasal spray were found to have no effect by Eby et al. in a 1984 unpublished report of a double-blind, placebo-controlled clinical trial of natural colds in Austin involving 77 patients. Zinc orotate is an insoluble, nonionizable compound at pH 7.4, having a first stability constant of log K1 = 6.42 at 25°C.(37) Because no Zn2+ ions were released at pH 7.4, ZIA was 0.
Nonastringent zinc aspartate (24 mg zinc) compressed lozenges and acetaminophen were reported to have no effect compared with placebo and acetaminophen in a double-blind, placebo-controlled, clinical trial of natural colds in 100 patients (personal communication, M L McCutcheon, M.D., Student Health Director, University of Minnesota at Duluth, 1987). Pleasant-tasting zinc aspartate has a first stability constant near log K1 = 5.9 at 37°C.(38) The conditional first stability constant at pH 6.8 is log K1 = 2.9 because of the protonated ammonium group.(39) Greatly confounding this analysis was the inclusion of 12 additional ingredients, most being water soluble and therefore potentially reactive with zinc according to the manufacturer (personal communication, B Schennum, National Sales Manager, Quantum, Eugene, OR, September 4, 1989). The high stability constant resulted in essentially no Zn2+ ion release at pH 7.4; therefore, ZIA was 0.
Non-astringent zinc gluconate (23 mg zinc) corn syrup- and sucrose-based, hard-boiled candy lozenges with 90 mg citric acid (1.3 moles citric acid to zinc) as flavor-mask were tested by Farr et al. in 1987 in 2 double-blind, placebo-controlled, clinical trials of clinically induced human rhinovirus-13 and -39 colds involving a total of 55 patients at several centers in the U.S. Compared with placebo, zinc gluconate-citrate lozenges worsened cold severity slightly.(35) The first stability constant of zinc citrate is log K1 = 4.7 at 37°C.(39,40) Zinc, weakly bound to gluconate, readily dissociates and binds strongly and instantly with citrate. The reaction leaves no Zn2+ ions at physiologic pH 7.4, and several negatively charged zinc-ligand complexes predominate.(39,40) Negatively charged zinc complexes are repelled from cell surfaces which are also negatively charged.(36) These lozenges lengthened colds by about one day.(35) Straight-line projection of nonnegative values in Table 2 and the increase in duration were used to estimate ZIA. The ZIA value was estimated to be negative 11 because of absence of Zn2+ ions, presence of negatively charged zinc species, and the 1 day increase in duration of colds.
Douglas et al. found effervescent mannitol-based lozenges containing 10 mg of zinc from zinc acetate to increase the average duration of colds by 4.4 days in a 1987 double-blind, placebo-controlled, clinical trial of natural colds in Australia involving 70 patients.(33) Effervescence always requires a chemical reaction, but the chemicals were not stated in the Douglas et al. report. Several hundred milligrams of tartaric acid and sodium bicarbonate were included to produce effervescence according to the lozenge manufacturer (personal communication, R. J. E. Williams, Development Scientist, F. H. Faulding and Company Limited, Adelaide, South Australia, August 27, 1987). Zinc acetate dissociates in solution releasing 100% of its zinc as Zn2+ ion.(41) Tartaric acid was present in considerable excess, and it has a high first stability constant for zinc at log K1 = 5.00.(42) Consequently, zinc binds strongly and preferentially to tartarate, and very few Zn2+ ions remain available at physiologic pH 7.4 while negatively charged zinc species predominate. Sodium bicarbonate reacts mainly with tartaric acid, but it may also react with zinc tartarate. Because of negatively charged zinc tartarate species and considerably worsened colds, the ZIA value for tartarate and carbonate complexed zinc acetate lozenges was estimated to be negative 55 by straight-line projection of non-negative values in Table 2 and the 4.4 day increase in duration of colds.
In a double-blind, placebo-controlled, clinical trial involving 87 patients by Godfrey et al. at Dartmouth College in 1992, zinc gluconate with glycine (10 mol glycine to 1 mol of zinc) as flavor-mask in 4.5-gram sucrose- and corn syrup-based, hard-boiled candy lozenges and acetaminophren reduced natural common cold "average duration" by 1.27 days compared with placebo and acetaminophren.(34) The reported average duration of colds appeared in actuality to be half-lives of colds. As a result, the authors' conclusions that (1) zinc gluconate-glycine lozenges shortened average duration of colds by 42%, and (2) placebos were efficacious by comparing their half-lives with historic average duration of colds appear methodologically erroneous.(34) The first stability constants for zinc glycinate complexes are slightly higher than for zinc citrate; for example log K1 = 4.8 at 37°C, and log K1 = 4.7 at 37°C,(39,40) respectively. With such an excess of glycine, zinc readily dissociates from gluconate and binds strongly to glycine. At the published salivary pH of 5.0 using these lozenges, available Zn2+ ion concentration is 20% according to computations by Berthon for zinc gluconate-glycine,(25) rather than 60% as shown in Figure 1. Zinc2+ ion concentration decreases to 0 in the zinc-gluconate-glycine system by pH 6.4, with only neutrally and negatively charged zinc glycinate species and traces of negatively charged zinc hydroxide existing at physiologic pH 7.4 accounting for some pseudo-astringency.(25) As would be expected from the closely related first stability constants, Zn2+ ion concentrations by pH resemble those for zinc gluconate and extramolar citric acid
In this anomaly from the ZIA - day reduction trend, ZIA is inestimable as the ZIA value is a horizontal line at 1.27 / ln2 days in Figure 2. The ZIA value is negative and would be found in the upper left quadrant of Figure 2 because of lack of Zn2+ ions, presence of mostly (85%) neutrally charged zinc glycinate and small amounts (15%) of negatively charged zinc glycinate species at physiologic pH 7.4, according to Berthon.(25) Inexplicably, Godfrey et al. claim 93% of the zinc was present as Zn2+ ion.(34)
ZIA factors from the preceding clinical trials are shown in Table 1. Zinc dosage is the total amount of zinc per lozenge. The fraction of zinc as ionic zinc is the fraction of total zinc available at pH 7.4 as solution Zn2+ ions. Dissolution times of lozenges are the average times required for lozenges to dissolve in the mouth. In every case, oral dissolution times of lozenges are considerably longer than dissolution times found using United States Pharmacopoeia (USP) disintegration tests. The term doses per day means the number of dosages actually taken, not the number of dosages planned to be taken. Total saliva is the total volume of saliva generated from use of lozenges. Lozenge weight is the actual weight of the lozenges tested.
ZIA values derived from the preceding clinical trials using analytical methods previously described are shown in Table 2. The table also shows concentration of Zn2+ ions at pH 7.4 (Zi), electronic charges of zinc species at pH 7.4, and treatment efficacy of lozenges. Complete details of ZIA calculations, previously unpublished manufacturers' lozenge formulations, methods and procedures of lozenge manufacture having considerable bearing upon lozenge ZIA values, and other data too complex or lengthy for presentation in this article have been published.(25)
Nonnegative Table 2 data were analyzed to determine the correlation between ZIA and reduction in duration of common colds as shown in Figure 2. Spearman's rank difference correlation method was used. The ZIA value (dose) and reduction in duration (response) were linearly related. The statistic was found to be rho = 0.96. The test of the hypothesis that the population correlation is zero is significant beyond the 0.02 level in a 2-tailed test.
One is warranted in rejecting the null hypothesis that there is no relationship between ZIA values and reductions in duration of common colds. A regression analysis of these data points yielded the equation y = 0.077 x - 0.16.
Lozenge safety was not considered to be an issue in any of the reanalyzed reports because zinc is generally regarded as a safe substance as well as an essential human nutrient. The US recommended daily allowance for zinc is 15 mg. Short-term administration of dosages used in common cold clinical trials does not change blood parameters or cause any lasting side effects. However, administration of similar doses for more than the 7-day treatment period is not recommended. Long-term administration of excess zinc can interfere with copper absorption resulting in reversible hemopoietic and immunologic adverse effects.(44)
None of the results of zinc lozenges for common cold trials included evidence of harmful effects. The main adverse effects in the 1984 trials by Eby et al. were minor oral irritation and objectionable aftertaste.(1,25) The pleasant-tasting, fructose-based lozenges used in the trial by Al-Nakib et al. did not cause any reported oral adverse effects, and they were reported to be well tolerated in spite of the large number used. There were no adverse changes in hematologic or biochemical indices, although zinc-treated patients all showed a marked increase in urinary zinc excretion.(2) Smith et al. reported no adverse effects from treatment except for strongly objectionable taste.(31) Weisman et al. reported no significant differences between zinc- and placebo-treated patients in side effects.(32) Douglas et al. reported no adverse effects other than oral irritation.(33) Godfrey et al. reported gastro-intestinal discomfort and minor oral adverse effects.(34) Farr et al. reported no significant differences between zinc- and placebo-treated patients in oral adverse effects, vital signs, hematologic parameters, metabolic profiles, and urinalyses.(35)
In nine unrelated, healthy volunteers, the present author observed an 80- to 120-mV potential difference between the interior of the mouth and the interior of the nose that appears to be evidence of a BCEC.(25) Placement of one electrode of an ohmmeter onto an oral cavity surface and the other onto a nasal turbinate usually showed either a 9,000- to 10,000-ohm reading or a reading greater than 15,000 ohms, dependent only upon electrode placement (polarity). The highest mouth-nose differential resistance readings observed were 50,000 and 100,000 ohms in an individual extremely resistant to upper respiratory infections. Reversal of electrodes always resulted in a significant change in resistance, mimicking a diode effect. No differences were found in BCEC differentials before, during, and for one hour after use of ZIA 100 lozenges. The author has not studied the BCEC in common cold patients.(25)