Foreword (key findings revisited)
tl;dr
The CDC and the FDA have not found a common causal agent in the cartridges/products used by all people who have developed the illness.
The presence of cadmium has not been confirmed yet in the products used by patients who developed EVALI. The hypothesis remains plausible:
Key findings from our vape pen study
Findings relate only to airflow-activated ("auto") vape pens, as only these designs have an open channel into the battery compartment. Our results indicate plausible scenarios for contamination of vape liquid and transfer of contaminated vapor to the lungs.
Quite unexpectedly, we also discovered the lithium ion cells contained within unbranded/counterfeit vape pens can be traced back to their manufacturer. This may be useful in determining the original source of the tainted hardware and cartridges.
Using the information printed on the lithium cells, we were able to identify the OEM of the batteries used in the unbranded vape pens. It may be possible to use this information to track assembled devices back to the manufacturer, following the information contained on the batteries alone.
Recommendations
Vape Pen Vivisection
To assess the exposure model, I used motive, means, and opportunity as a framework again. I will expand on some of the previously introduced concepts and include visual evidence to support.
Cheap Manufacturing and/or Cadmium Solder ("motive"). The causal agent or contaminant must be present in the vape pen hardware for a valid reason, and that reason is money.
Cadmium-containing silver solder is less expensive than cadmium-free alternatives, and has improved flow properties which facilitate the joining of dissimilar metals found in vape pen electrical components.
Expanding on this concept: Regardless of cadmium content, not all vape pens are created equal. Clean and safe manufacturing processes are always more expensive than quick and dirty assembly using crude, exposed solder joints.
Both of the examples below come from auto vape pens and are in direct contact with vape liquid. In assembled vape pens, these components are located directly below the atomizer and would be exposed to any liquid that drips from the heating coil into the battery area.
Contamination would be minimal in the example on the left. The leads are completely insulated, preventing short-circuit and electrical shock, and there are no exposed globules of metal solder to leach into liquid that falls into the compartment.
The example on the right is functional, but....less correct. Dropping 40 mg blobs of solder onto 2mm wires requires very little expertise or training, allows rapid turnover of assembled and working units, uses cheap solder and skips the extra step of insulating the exposed metal.
Bathing this amount of solder in vape liquid would unquestionably expose the user to significant amounts of aerosolized metal, and that would present a clear health risk. The type of solder used would determine the severity of illness. If I'm correct about the cadmium, it will most likely appear in hardware depicted below and to the right.
tl;dr
- Inhalation-activated vape pens expose users to the contents of the battery compartment
- Cheap cartridges leak into the battery compartment, and may leach metal from exposed solder
- Liquid and vapor contaminated with heavy metal can be inhaled by vape pen users
The CDC and the FDA have not found a common causal agent in the cartridges/products used by all people who have developed the illness.
The presence of cadmium has not been confirmed yet in the products used by patients who developed EVALI. The hypothesis remains plausible:
- Progression of cadmium pneumonitis (Greenberg & Vearrier, 2015) closely resembles the clinical findings of patients with EVALI (Layden et al., 2019; Schier et al., 2019)
- Cadmium is frequently present in solder used in the construction of cheap electronic devices like vape pens
- When they are present in vape pens or e-liquid, cadmium and other toxic metals have been shown to transfer into the lungs on vapor (Aherrera et al., 2017; Goniewicz et al., 2018)
Key findings from our vape pen study
Findings relate only to airflow-activated ("auto") vape pens, as only these designs have an open channel into the battery compartment. Our results indicate plausible scenarios for contamination of vape liquid and transfer of contaminated vapor to the lungs.
Quite unexpectedly, we also discovered the lithium ion cells contained within unbranded/counterfeit vape pens can be traced back to their manufacturer. This may be useful in determining the original source of the tainted hardware and cartridges.
- Significant Potential for Contamination. Vape liquid readily seeps into the battery compartment of airflow-activated ("auto") vape pens
- Heavy Metal Leaching. In the majority of auto vape pens examined, significant amounts (40mg-80mg) of exposed metal solder is in direct contact with vape liquid entering the battery compartment
- Significant Potential for Transfer to Lungs. Vape liquid that has entered the battery compartment can re-enter the atomizer compartment and be inhaled
- Traceability. The exterior of counterfeit vape pen hardware and cartridges contain no information on the identity of the manufacturer. Once disassembled, however, the lithium ion cells from the interior of the vape pens were are imprinted with 1) date of production, 2) barcode and battery model/product number, and, 3) the manufacturer's ID code.
Using the information printed on the lithium cells, we were able to identify the OEM of the batteries used in the unbranded vape pens. It may be possible to use this information to track assembled devices back to the manufacturer, following the information contained on the batteries alone.
Recommendations
- Examine the internal battery casing for signs of corrosion (ex. verdigris), as well as exposed metal solder.
- If corrosion and/or exposed solder are evident in the battery compartment, this may indicate vape liquid was exposed to heavy metal contamination. Exposed circuitry and solder joints in the battery compartment should be tested for a panel of toxic metals, including cadmium.
- If cadmium or other toxic metals are found in the battery compartment, whole blood samples should be evaluated to establish the level of acute exposure. Long term monitoring may be necessary, as cadmium has a urinary excretion half-life of 10-30 years. Long after the respiratory illness resolves, patients with significant exposure would be at increased risk of kidney damage (.
Vape Pen Vivisection
To assess the exposure model, I used motive, means, and opportunity as a framework again. I will expand on some of the previously introduced concepts and include visual evidence to support.
Cheap Manufacturing and/or Cadmium Solder ("motive"). The causal agent or contaminant must be present in the vape pen hardware for a valid reason, and that reason is money.
Cadmium-containing silver solder is less expensive than cadmium-free alternatives, and has improved flow properties which facilitate the joining of dissimilar metals found in vape pen electrical components.
Expanding on this concept: Regardless of cadmium content, not all vape pens are created equal. Clean and safe manufacturing processes are always more expensive than quick and dirty assembly using crude, exposed solder joints.
Both of the examples below come from auto vape pens and are in direct contact with vape liquid. In assembled vape pens, these components are located directly below the atomizer and would be exposed to any liquid that drips from the heating coil into the battery area.
Contamination would be minimal in the example on the left. The leads are completely insulated, preventing short-circuit and electrical shock, and there are no exposed globules of metal solder to leach into liquid that falls into the compartment.
The example on the right is functional, but....less correct. Dropping 40 mg blobs of solder onto 2mm wires requires very little expertise or training, allows rapid turnover of assembled and working units, uses cheap solder and skips the extra step of insulating the exposed metal.
Bathing this amount of solder in vape liquid would unquestionably expose the user to significant amounts of aerosolized metal, and that would present a clear health risk. The type of solder used would determine the severity of illness. If I'm correct about the cadmium, it will most likely appear in hardware depicted below and to the right.
Not All Vape Pens ("opportunity")
The opportunity for significant exposure to the battery compartment only occurs in "buttonless" or "auto" vape devices activated by inhalation.
Auto vape devices are commonly found as 2-piece reusable vape pens (rechargeable battery base used with disposable or refillable cartridges), or 1-piece disposable pens (e-cigarette design, single-use units).
The inhalation activation mechanism of auto vape devices exposes the user to the contents of the battery compartment via the air channel located in auto devices. This channel is not required for activation of vape pens with an external power button, and is therefore not present in these types of vape devices.
Figure 1: External button vape pen (left), Auto vape pen (right), with air channel into battery compartment
Not All Cartridges Are Equal ("means")
Cheap cartridges with low design complexity typically have just two vent ports and a disk-shaped flange incorporated into the cartridge base. As noted in a recent patent (Levitz, 2018), these features contribute to poorly controlled air intake and exhaust pressure in auto vape pens and leakage of vape liquid into the battery compartment inevitably occurs. Liquid in the battery compartment does not cause the device to immediately fail, however, and contaminated liquid can pass back into the atomizer of the vape device, convert to vapor, and be inhaled.
From United States patent (US 9,675,107 B2), "Battery Connector for Electronic Cigarette With Side Air Intake", https://patentimages.storage.googleapis.com/3c/e8/cb/38d052d5aebbfd/US9675107.pdf.
"there has often been a specific problem with [auto vape pens]. When the smoker exhales into a [auto vape pen] (as smokers often do), the vapors from the mouth follow the reverse of the path described above - they flow inside the Battery towards the sensor and often condense causing the sensor to become clogged and fail. The exhalation likewise causes excess liquid in the Cartomizer to follow the same path into the Battery part towards the sensor.".
Figure 2: Cartridge Features
Dank Carts and counterfeit "Cookies" brand cartridges have only two (2) air intake/exhaust vents and flattened disk flanges in cartridge thread. Cartridges with cross-cut flanges similar to those described in Levitz patent have better air flow and decreased risk of leakage into battery area.
Cheap cartridges with low design complexity typically have just two vent ports and a disk-shaped flange incorporated into the cartridge base. As noted in a recent patent (Levitz, 2018), these features contribute to poorly controlled air intake and exhaust pressure in auto vape pens and leakage of vape liquid into the battery compartment inevitably occurs. Liquid in the battery compartment does not cause the device to immediately fail, however, and contaminated liquid can pass back into the atomizer of the vape device, convert to vapor, and be inhaled.
From United States patent (US 9,675,107 B2), "Battery Connector for Electronic Cigarette With Side Air Intake", https://patentimages.storage.googleapis.com/3c/e8/cb/38d052d5aebbfd/US9675107.pdf.
"there has often been a specific problem with [auto vape pens]. When the smoker exhales into a [auto vape pen] (as smokers often do), the vapors from the mouth follow the reverse of the path described above - they flow inside the Battery towards the sensor and often condense causing the sensor to become clogged and fail. The exhalation likewise causes excess liquid in the Cartomizer to follow the same path into the Battery part towards the sensor.".
Figure 2: Cartridge Features
Dank Carts and counterfeit "Cookies" brand cartridges have only two (2) air intake/exhaust vents and flattened disk flanges in cartridge thread. Cartridges with cross-cut flanges similar to those described in Levitz patent have better air flow and decreased risk of leakage into battery area.
Figure 3: Exposure to the Battery Compartment in Airflow-Activated Vape Devices ("means and opportunity" for inhalation exposure)
Inhalation (left). Air is drawn through sensor in battery compartment, activating the heating coil in the atomizer. Vapor produced from heating coil is inhaled.
Exhalation (middle). On exhale or release of suction, vapor and liquid may be push/pulled into the battery compartment and exposed to metal solder and other contaminants.
Re-Inhalation (right) Contaminated vape liquid can be drawn back into cartridge atomizer, vaporized, and inhaled
Inhalation (left). Air is drawn through sensor in battery compartment, activating the heating coil in the atomizer. Vapor produced from heating coil is inhaled.
Exhalation (middle). On exhale or release of suction, vapor and liquid may be push/pulled into the battery compartment and exposed to metal solder and other contaminants.
Re-Inhalation (right) Contaminated vape liquid can be drawn back into cartridge atomizer, vaporized, and inhaled
Considerable Risk of Exposure to Heavy Metal in Some Inhalation-Activated Vape Pens ("means")
40mg to 60mg of exposed metal solder was present in some of the models of vape pens evaluated in our study.
40mg to 60mg of exposed metal solder was present in some of the models of vape pens evaluated in our study.
High Variability of Exposed Solder Levels
Identical units (left, three (3) unit of Model A, and right, three (3) units of Model B) same model) exhibited high degree of variability in the amount of exposed solder inside the battery compartment.
These are not visible unless the vape pen is completely disassembled. The variability in solder quantity may provide insight to explain two contrasting traits of the EVALI epidemic: wide geographical dispersion and low number of affected people. Wide dispersion suggests a large production batch, whereas the limited number of EVALI patients indicates this illness is constrained somehow. Variable levels of contaminated solder would correlate with varying degrees of exposure to toxic contaminants, and sufficient exposure to cause EVALI may only occur in certain circumstances (leaking cartridges, heavy/daily vape use, etc.).
Vaping Type vs Vaping Frequency: Utah
A recent report on the EVALI outbreak in Utah tracked the type of vaping (nicotine or THC) and also the frequency of vaping per day (Lewis N, 2019). The tables below are adapted from that report.
The largest group of patients (60%) used both nicotine- and THC-containing products, which is consistent with the national findings.
The relationship between frequency of vaping and type was interesting.
Identical units (left, three (3) unit of Model A, and right, three (3) units of Model B) same model) exhibited high degree of variability in the amount of exposed solder inside the battery compartment.
These are not visible unless the vape pen is completely disassembled. The variability in solder quantity may provide insight to explain two contrasting traits of the EVALI epidemic: wide geographical dispersion and low number of affected people. Wide dispersion suggests a large production batch, whereas the limited number of EVALI patients indicates this illness is constrained somehow. Variable levels of contaminated solder would correlate with varying degrees of exposure to toxic contaminants, and sufficient exposure to cause EVALI may only occur in certain circumstances (leaking cartridges, heavy/daily vape use, etc.).
Vaping Type vs Vaping Frequency: Utah
A recent report on the EVALI outbreak in Utah tracked the type of vaping (nicotine or THC) and also the frequency of vaping per day (Lewis N, 2019). The tables below are adapted from that report.
The largest group of patients (60%) used both nicotine- and THC-containing products, which is consistent with the national findings.
The relationship between frequency of vaping and type was interesting.
- The largest group of patients (55%) reported high frequency (> 25 times per day) of vaping nicotine products
- A similar percentage of patients (54%) reported only modest use (1-5 times per day) of THC products
Self-reported product use behaviors in the 3 months before injury onset in interviewed patients with electronic cigarette (e-cigarette), or vaping, product use–associated lung injury (N = 53) — Utah, April– October 2019 Product use and behavior |
% of patients |
THC-containing product use |
Value |
Any use |
92% |
Exclusive use |
32% |
Nicotine-containing product use |
Value |
Any use |
66% |
Exclusive use |
6% |
Both nicotine- and THC-containing product use |
60% |
Characteristics of tetrahydrocannabinol (THC)- or nicotine-containing products used in the 3 months preceding illness onset in patients with electronic cigarette (e-cigarette), or vaping, product use–associated lung injury (N = 131) — Utah, April–October 2019
Frequency of use (times per day) |
THC-containing products |
Nicotine-containing products |
< 1 |
11% |
10% |
1 - 5 |
54% |
17% |
6 - 25 |
10% |
17% |
> 25 |
24% |
55% |
Visual Evidence of Liquid Entering Battery Compartment in Auto Vape Pens ("opportunity")
The presence of visible oxidized copper scale (verdigris) inside the battery compartment is a positive indicator that vape liquid readily enters the battery compartment. Worth mentioning again that this only occurs in auto vape pens.
Oxidation was variable in units examined. Minimal in some cases (left), very evident in others (middle), and not evident in some cases (right).
The presence of visible oxidized copper scale (verdigris) inside the battery compartment is a positive indicator that vape liquid readily enters the battery compartment. Worth mentioning again that this only occurs in auto vape pens.
Oxidation was variable in units examined. Minimal in some cases (left), very evident in others (middle), and not evident in some cases (right).
Traceability
None of the hardware (cartridges or auto pens) examined in this study contained branding information or serial numbers. The lithium polymer batteries contained within the vape pens, however, are imprinted with a model number, a two- to three-digit manufacturer's code, the production date, and often a barcode with additional information. Using Google or Alibaba, we were able to use this information to determine the manufacturer of the battery. Further work would be required to use that information to source the assembled vape pens, but it should be possible.
None of the hardware (cartridges or auto pens) examined in this study contained branding information or serial numbers. The lithium polymer batteries contained within the vape pens, however, are imprinted with a model number, a two- to three-digit manufacturer's code, the production date, and often a barcode with additional information. Using Google or Alibaba, we were able to use this information to determine the manufacturer of the battery. Further work would be required to use that information to source the assembled vape pens, but it should be possible.
Conclusions
Auto vape devices, which include the 2-piece auto vape pens examined here and 1-piece disposable e-cigarettes, expose users to the contents of the battery compartment.
Motive. In well-manufactured devices, this may not be cause for serious concern, but there is financial motive for manufacturers to mass produce cheap vape devices. In the low quality devices examined here, varying, but significant, amounts of metal solder are present in the flowpath leading to the lungs.
Means and Opportunity. Bidirectional transfer of vape liquid between the cartridge and battery compartment is possible, and likely. Contamination of vape liquid with components of the solder is also probable. Depending on the type of solder used, heavy metals of varying toxicity would be inhaled by the user.
These types of devices are ubiquitous in the vape pen market. If similar hardware is linked to patients who have developed the illness, solder exposure represent a plausible explanation for EVALI and should be investigated.
Auto vape devices, which include the 2-piece auto vape pens examined here and 1-piece disposable e-cigarettes, expose users to the contents of the battery compartment.
Motive. In well-manufactured devices, this may not be cause for serious concern, but there is financial motive for manufacturers to mass produce cheap vape devices. In the low quality devices examined here, varying, but significant, amounts of metal solder are present in the flowpath leading to the lungs.
Means and Opportunity. Bidirectional transfer of vape liquid between the cartridge and battery compartment is possible, and likely. Contamination of vape liquid with components of the solder is also probable. Depending on the type of solder used, heavy metals of varying toxicity would be inhaled by the user.
These types of devices are ubiquitous in the vape pen market. If similar hardware is linked to patients who have developed the illness, solder exposure represent a plausible explanation for EVALI and should be investigated.
References
Aherrera, A., Olmedo, P., Grau-Perez, M., Tanda, S., Goessler, W., Jarmul, S., ....Navas-Acien, A. (2017). The association of e-cigarette use with exposure to nickel and chromium: A preliminary study of non-invasive biomarkers. Environ Res, 159, 313-320. doi: 10.1016/j.envres.2017.08.014
Bernhoft, R. A. (2013). Cadmium toxicity and treatment. ScientificWorldJournal, 2013, 394652. doi: 10.1155/2013/394652
Goniewicz, M. L., Smith, D. M., Edwards, K. C., Blount, B. C., Caldwell, K. L., Feng, J., . . . Hyland, A. J. (2018). Comparison of Nicotine and Toxicant Exposure in Users of Electronic Cigarettes and Combustible Cigarettes. JAMA Netw Open, 1(8), e185937. doi: 10.1001/jamanetworkopen.2018.5937
Greenberg, M. I., & Vearrier, D. (2015). Metal fume fever and polymer fume fever. Clin Toxicol (Phila), 53(4), 195-203. doi: 10.3109/15563650.2015.1013548
Layden, J. E., Ghinai, I., Pray, I., Kimball, A., Layer, M., Tenforde, M., . . . Meiman, J. (2019). Pulmonary Illness Related to E-Cigarette Use in Illinois and Wisconsin - Preliminary Report. N Engl J Med. doi: 10.1056/NEJMoa1911614
Lewis N, M. K., Sage K, et al. (2019). E-cigarette Use, or Vaping, Practices and Characteristics Among Persons with Associated Lung Injury — Utah, April–October 2019. Morb Mortal Wkly Rep, 68, 953-956. doi: http://dx.doi.org/10.15585/mmwr.mm6842e1
Schier, J. G., Meiman, J. G., Layden, J., Mikosz, C. A., VanFrank, B., King, B. A., . . . Meaney-Delman, D. (2019). Severe Pulmonary Disease Associated with Electronic-Cigarette-Product Use - Interim Guidance. MMWR Morb Mortal Wkly Rep, 68(36), 787-790. doi: 10.15585/mmwr.mm6836e2
Williams, M., Villarreal, A., Bozhilov, K., Lin, S., & Talbot, P. (2013). Metal and silicate particles including nanoparticles are present in electronic cigarette cartomizer fluid and aerosol. PLoS One, 8(3), e57987. doi: 10.1371/journal.pone.0057987
Aherrera, A., Olmedo, P., Grau-Perez, M., Tanda, S., Goessler, W., Jarmul, S., ....Navas-Acien, A. (2017). The association of e-cigarette use with exposure to nickel and chromium: A preliminary study of non-invasive biomarkers. Environ Res, 159, 313-320. doi: 10.1016/j.envres.2017.08.014
Bernhoft, R. A. (2013). Cadmium toxicity and treatment. ScientificWorldJournal, 2013, 394652. doi: 10.1155/2013/394652
Goniewicz, M. L., Smith, D. M., Edwards, K. C., Blount, B. C., Caldwell, K. L., Feng, J., . . . Hyland, A. J. (2018). Comparison of Nicotine and Toxicant Exposure in Users of Electronic Cigarettes and Combustible Cigarettes. JAMA Netw Open, 1(8), e185937. doi: 10.1001/jamanetworkopen.2018.5937
Greenberg, M. I., & Vearrier, D. (2015). Metal fume fever and polymer fume fever. Clin Toxicol (Phila), 53(4), 195-203. doi: 10.3109/15563650.2015.1013548
Layden, J. E., Ghinai, I., Pray, I., Kimball, A., Layer, M., Tenforde, M., . . . Meiman, J. (2019). Pulmonary Illness Related to E-Cigarette Use in Illinois and Wisconsin - Preliminary Report. N Engl J Med. doi: 10.1056/NEJMoa1911614
Lewis N, M. K., Sage K, et al. (2019). E-cigarette Use, or Vaping, Practices and Characteristics Among Persons with Associated Lung Injury — Utah, April–October 2019. Morb Mortal Wkly Rep, 68, 953-956. doi: http://dx.doi.org/10.15585/mmwr.mm6842e1
Schier, J. G., Meiman, J. G., Layden, J., Mikosz, C. A., VanFrank, B., King, B. A., . . . Meaney-Delman, D. (2019). Severe Pulmonary Disease Associated with Electronic-Cigarette-Product Use - Interim Guidance. MMWR Morb Mortal Wkly Rep, 68(36), 787-790. doi: 10.15585/mmwr.mm6836e2
Williams, M., Villarreal, A., Bozhilov, K., Lin, S., & Talbot, P. (2013). Metal and silicate particles including nanoparticles are present in electronic cigarette cartomizer fluid and aerosol. PLoS One, 8(3), e57987. doi: 10.1371/journal.pone.0057987