Potential harmful health effects of inhaling nicotine-free shisha-pen vapor: a chemical risk assessment of the main components propylene glycol and glycerol

A shisha-pen is an electronic inhaler that vaporises a liquid solution consisting
mainly of humectants and flavors into an aerosol mist. Like e-cigarettes, shisha-pens
simulate the act of tobacco smoking. In the present study, the contents and vapor
of shisha-pens without nicotine with different flavors (apple, strawberry, grape)
were analyzed.

The major components found in the liquid of shisha-pens were propylene glycol and
glycerol (54%/46%). The manufacturer reports a minimum of 500 puffs to be taken from
one shisha-pen (shisha-pen package). This was confirmed by our smoking machine analysis,
in which up to 630 puffs were taken from one shisha-pen. GC analysis of shisha-pen
puffs showed that the vapor in the shisha-pen was comprised of an average of 0.7 mg/puff
of propylene glycol and 0.6 mg/puff of glycerol. In addition, the vapor contained
a small amount of flavor and other trace components (1%). No tri-ethyleneglycol,
di-ethyleneglycol and nicotine were found. Furthermore, pyrolysis of shisha-pen vapor
did not show presence of well-known tobacco smoke components, such as benzene or 1,3-butadiene.

Risk assessment was performed per major component found in shisha-pen vapor, propylene
glycol and glycerol. The maximum concentration of propylene glycol and glycerol that
would reach the lower respiratory tract after one puff was estimated, as described
previously 3],4]. For the shisha-pen, the amount of puffs taken per unit time, the volume of vapor
inhaled and the length of vaping sessions of the average shisha-pen user remain unknown.
For this reason, smoking topography described for use of “normal” cigarettes 5] and e-cigarettes 8],9],6] were used. Calculations were made for a 1-puff scenario.

Propylene glycol is used in the food, cosmetic, pharmaceutical and plastic industries.
It is also commonly used to create the artificial smoke or mist often seen in discotheques,
theatre and television productions 10]. Glycerol is widely used in many industrial and consumer products, e.g. soaps/detergents,
medicines, cosmetics, food, drinks, paints, resins and paper 10]. Both substances are “generally recognized as safe” (GRAS) for use as food additives
11]. The GRAS approval, however, does not apply to exposure to propylene glycol and glycerol
through the shisha-pen. This is because in this scenario the substances are not ingested
as in food, but inhaled, which results in exposure of the respiratory tract and lungs.
For propylene glycol, it is known that repeated, short-term exposure of eyes, skin,
nose, and mouth may cause irritation 12].

The concentrations that reached the airways and lungs after using a shisha-pen were
compared to data from published toxicity studies 13]-15]. Studies were selected based on resemblance of the exposure scenario to that of shisha-pen
use. Differences between studies and the actual exposure to shisha-pen use, such as
differences in duration of exposure and differences between animals and humans, were
taken into account when only animal studies were available.

Hazard assessment of propylene glycol showed that there is no evidence that propylene
glycol is carcinogenic to humans (The Health Council of the Netherlands 16]). Non-carcinogenic, local respiratory effects and systemic effects following propylene
glycol exposure showed an increased number of goblet cells in the respiratory tract
and nasal hemorrhaging observed when rats were exposed to 160 mg/m
³
(the lowest concentration tested), 6 hours per day, 5 days per week for 13 weeks 14]. Effects such as nasal burning, stinging and throat irritation were attributed to
exposure to propylene glycol as part of a pharmaceutical formulation inhaled by patients
suffering from allergic rhinitis for 4 weeks. However, these effects were significantly
less following a change in the content of propylene glycol in the formulation from
20% to 5% 17]. Furthermore, acute ocular and upper airway irritation was caused by short exposure
to propylene glycol mist from artificial smoke generators in non-asthmatic human volunteers
(n?=?27) who were exposed in an aircraft simulator to propylene glycol mist for 1 minute.
A few (4 out of 27) reacted with cough and slight airway obstruction 15]. Minor systemic effects were observed only in female rats which included body weight
reduction and changes in leukocyte profile. These systemic effects on body weight
and leukocyte profile have not been found consistently in other studies indicating
that gender differences in susceptibility to propylene glycol’s adverse effects in
the rat, but other studies do not provide additional evidence for this 17].

For risk assessment of propylene glycol the maximum alveolar concentrations in after
one puff was estimated to be 430 to 603. The study of human volunteers (n?=?27) exposed
to propylene glycol for one minute at concentrations ranging from 176–851 mg/m
³
showed upper airway irritation 15]. It is not clear if irreversible effects will occur after prolonged use but an animal
study showed that repeated exposure (6 h per day; 5 days per week) for 90 days at
1000 and 2200 mg/m
³
caused irreversible respiratory damage 14]. Limits for propylene glycol by actors exposed via theatrical fog has been set at
40 mg/m
³18]. The estimated maximum alveolar concentration of propylene glycol in one puff exceeds
this peak acceptable concentration. This analysis of the shisha-pen demonstrates that
a risk of irritating effects on the respiratory tract epithelium due to propylene
glycol exists. Details on risk assessment of propylene glycol (exposure assessment,
PoD, and risk on local effects) is presented in Risk assessment propylene glycol;
1-puff scenario section (propylene glycol; 1-puff scenario). The MOE analysis is presented
in Table 1.

Table 1. Summary MOE analysis, propylene glycol, 1-puff scenario, human study used as PoD

Risk assessment propylene glycol; 1-puff scenario

Step 1: Exposure assessment

For the exposure scenario, the same method as previously described 7] was utilized with a few adaptations. Puffing patterns (puff frequency, strength and
duration) vary considerably among individuals who smoke electronic cigarettes or shisha-pens,
but surveys indicate that individuals take an average of 120–175 puffs per day 8],9]. There is no data available on the duration of shisha-pen smoking sessions and therefore
we can only assume that 1 puff has a volume of 50 mL as it is with cigarette smoke
5], or 70 mL as is found with e-cigarettes 6]. We must keep in mind that with the nicotine-free shisha-pen, the user will not adjust
the volume to satisfy the nicotine craving, for this reason we used both volumes to
obtain a range of exposure as an indication of the overall risk.

The average concentration per shisha-pen smoking session can be calculated by adapting
the exposure scenario described previously for cigarette smoking 7] and dividing the amount in mg inhaled during a shisha-pen session (D _{1-puff shisha-pen}
) by 0.05 L
¹
, or 0.07 L
²
.

¹
C
_alv;max
?=?0.042 × D _{1-puff shisha-pen}
/0.05?=?0.85 × D _{1-puff shisha-pen}
?=?mg/L

²
C
_alv;max
?=?0.042 × D _{1-puff shisha-pen}
/0.07?=?0.6 × D _{1-puff shisha-pen}
?=?mg/L

GC analysis showed that the smoke in the shisha pen was comprised of an average of
0.71 mg/puff of propylene glycol:

¹
C
_alv;max
?=?0.85 × D _{1-puff shisha-pen}
?=?0.85 × 0.71 mg?=?0.603 mg/L?=?603 mg/m
³

²
C
_alv;max
?=?0.85 × D _{1-puff shisha-pen}
?=?0.6 × 0.71 mg?=?0.43 mg/L?=?430 mg/m
³

The estimated inhaled concentration of propylene glycol per puff was 0.71 mg with
a maximum alveolar concentration (C _alv;max ) of 430 to 603 mg/m
³
.

Step 2: Point of departure

One human study in which humans were exposed to an aerosol mist as part of an aviation
emergency training was considered the best PoD for further risk assessment. Please
refer to Table 1 for MOE calculation.

Step 3: Risk on local effects

The MOE for respiratory tract irritation was found to range from 0.3 to 2 (Table 1). Considering the MOE, a risk of effects on the respiratory tract epithelium due
to propylene glycol exists. For evaluation of this MOE it needs to be taken into account
that the lowest observed adverse effect level (LOAEL) was used as PoD instead of no
observed adverse effect level (NOAEL).

Hazard assessment of glycerol showed no evidence for carcinogenic effects. Non-carcinogenic,
local respiratory and systemic effects were reported as local irritant effects to
the upper respiratory tract observed when rats were exposed to 662 mg/m
³
, 6 hours per day, 5 days per week for 13 weeks, with no toxic effects observed at
165 mg/m
³19]. No systemic effects were reported in this study or in a study with rats exposed
to concentrations of 1000, 1930 and 3910 mg/m
³
, 6 hours per day, 5 days per week for 14 days 19].

For risk assessment of glycerol, the maximum alveolar concentration of glycerol after
one puff was estimated to be 348 to 495 mg/m
³
. Due to lack of relevant human inhalation studies with glycerol, no MOE was calculated.
Nevertheless, two animal studies showed that continuous exposure (6 h per day; 5 days
per week) for 14 and 90 days showed irritation to the upper respiratory tract at 662
and 1000 mg/m
³
, respectively 19]. Given the high inhaled concentration of glycerol in one puff, a risk of irritating
effects on the respiratory tract epithelium due to glycerol exists with increased
duration of shisha-pen exposure. Details on risk assessment of glycerol (exposure
assessment, PoD, and risk on local effects) is presented in Risk assessment glycerol;
1-puff scenario section (glycerol; 1-puff scenario).

Risk assessment glycerol; 1-puff scenario

Step 1: Exposure assessment

For the exposure scenario, the same method as previously described 7] was utilized with a few adaptations. Puffing patterns (puff frequency, strength and
duration) vary considerably among individuals who smoke electronic cigarettes or shisha-pens,
but surveys indicate that individuals take an average of 120–175 puffs per day 9],8]. There is no data available on the duration of shisha-pen smoking sessions and therefore
we can only assume that 1 puff has a volume of 50 mL 5], or 70 mL 6]).

The average concentration per shisha-pen smoking session can be calculated by adapting
the exposure scenario described previously for cigarette smoking 7] and dividing the amount in mg inhaled during a shisha-pen session (D _{1-puff shisha-pen}
) ) by 0.05 L
¹
, or 0.07 L
²
.

¹
C
_alv;max
?=?0.042 × D _{1-puff shisha-pen}
/0.05?=?0.85 × D _{1-puff shisha-pen}
?=?mg/L

²
C
_alv;max
?=?0.042 × D _{1-puff shisha-pen}
/0.07?=?0.6 × D _{1-puff shisha-pen}
?=?mg/L

GC analysis showed that the smoke in the shisha pen was comprised of an average of
0.582 mg/puff of glycerol

¹
C
_alv;max
?=?0.85 × D _{1-puff shisha-pen}
?=?0.85 × 0.58 mg?=?0.495 mg/L?=?495 mg/m
³

²
C
_alv;max
?=?0.6 × D _{1-puff shisha-pen}
?=?0.6 × 0.58 mg?=?0.348 mg/L?=?348 mg/m
³

The estimated inhaled concentration of glycerol per puff was 0.58 mg with a maximum
alveolar concentration (C _alv;max ) of 348 to 495 mg/m
³
.

Step 2: Point of departure

Two studies with continuous exposure were found. The first had an NOAEL of 165 mg/m
³
and a LOAEL of 662 mg/m
³
for local irritant effect to the respiratory tract in rats exposed 6 h per day, 5 days
per week for 13 weeks (concentrations tested were 0, 33, 165 and 662 mg/m
³
) 19]. Another study showed an LOAEL of 1000 mg/m
³
for local irritant effects of the upper respiratory tract in rats exposed 6 h per
day, 5 days per week for 2 weeks (concentrations tested were 0, 1000, 1930 and 3910 mg/m
³
) 19]. It must be kept in mind that in the rat study, animals were exposed to glycerol
for 6 h per day and that these data were compared with 1 puff of a shisha-pen.

Step 3: Risk on local effects

Because a relevant study with a similar exposure pattern as that of a shisha-pen could
not be found, a reliable MOE could not be calculated. Nevertheless, the inhaled concentration
of glycerol in one puff was estimated to be 348 to 495 mg/m
³
, in comparison to an NOAEL of 165 mg/m
³
, and an LOAEL of 1000 and 662 mg/m
³
observed for local irritant effect in 2- and 13-week rat studies, respectively. Given
the high inhaled concentration of glycerol in one puff, a risk of irritating effects
on the respiratory tract epithelium due to glycerol exists with increased duration
of exposure.

For the risk assessments performed for propylene glycol and glycerol present in the
vapor of the shisha-pen, it is recognised that several assumptions have been made
and that the risk assessment can be refined reconsidering these assumptions. Although
such a refinement is beyond the scope of the present analysis, considering the low
MOE, it remains to be seen if further refinement will alter the conclusion.

The current study is the first to present a chemical analysis and subsequent assessment
of the risks of inhaling nicotine-free shisha-pen vapor, focusing on the major components
propylene glycol and glycerol. For propylene glycol risk assessment could rely on
a relevant human study, allowing for MOE analysis. Also, for glycerol animal data
were available allowing for estimation of risks upon exposure. Some limitations include
the lack of information on shisha-pen use; we can only assume that topography including
puff volume is within the range of that of cigarettes and e-cigarettes. Furthermore,
there were no human (propylene glycol) and animal (propylene glycol, glycerol) studies
that mimicked the exposure scenario of shisha-pen smoking. Other factors to be taken
into account when performing risk assessment include less-than-lifetime exposure,
interspecies extrapolation (rat to humans), and inter-individual variability. Moreover,
the current chemical risk assessment approach presents a single-component analysis
and the combined effects of propylene glycol and glycerol need further investigation.