E-cigarette explosions: patient profiles, injury patterns, clinical management, and outcome

Abstract

Background

E-cigarette abuse, also known as vaping, is a widespread habit. Recently, there have been increasing reports of explosions of these devices resulting in serious burn injuries, especially to the groin, hand, and face. Overheating rechargeable lithium-ion batteries are supposed to be the responsible mechanism, especially in low quality fabrications.

Methods

In this single-center retrospective study, data of 46 patients are presented and analyzed. In addition to information on demographics, injury patterns, and treatment options, this is the first study investigating outcome after an average of 13 months via a standardized telephone interview of 31 patients (67%). Patients were specifically asked regarding their outcome, vaping habits, technical modifications to their devices and supply source.

Results

Patients were mainly male (98.2%) with a median age of 36 years. Typical injury patterns included the groin region (n = 32; 69%), hands (n = 12; 25%) and face (n = 3; 7%). All patients underwent debridement, in nine cases hydrotherapeutically. 61% (n = 28) underwent consecutive tangential necrectomy and subsequent split thickness skin graft transplantation. Wound infection was observed in 18 patients (39%), with burn depth as a significant risk factor (p < 0.001). 91% of the followed-up patients were satisfied. Surprisingly, 38% were still using e-cigarettes. 42% (n = 13) reported manual modifications of their devices to prolong battery life or increasing smoke production.

Conclusions

Injuries from exploding e-cigarettes can be serious and should be treated in a specialized burn center. E-cigarette explosions lead to characteristic injury patterns and often need surgical treatment. This should be made more public to reduce their use and keep people from modifying the devices. The counterintuitive and irrational observation of a high rate of abuse even after the injury underlines their addiction potential.

Introduction

The abuse of tobacco and other smokable substances is known to mankind since its early days. Mainly due to the obvious health risks of cigarette smoking, the development and spread of electronic vaporing devices, also known as electronic nicotine delivery systems (ENDS), started approximately 15 years ago. Currently, ENDS are widely used. It has been shown that, especially in younger people, use of ENDS has already overrun the amount of combustion cigarette abuse.¹ These so called “e-cigarettes” are handheld devices, which are mostly driven by a rechargeable lithium-ion battery and need to be filled with liquids containing e.g., nicotine, alcohol, or aromatics. The prototype construction is shown in Fig. 1.²

Figure 1.

Prototype construction of an Electronic Nicotine Delivery System (ENDS). The ENDS is composed of a heater as aerosol generator, gauze saturated with e-liquid, a microprocessor (optional) to control operations, and an LED (optional) to imitate a burning coal. The energy source is a lithium-ion battery.² (Source: Figure was designed by Brown et al.² and used under the Creative Commons Attribution Non Commercial (CC BY-NC 3.0) license)

Recently, there is a growing number of reported cases in which an ENDS exploded, leading to serious injuries to the user.3, 4, 5, 6, 7, 8, 9 The lithium-ion batteries are considered the main culprit for these events. Conversely, lithium-ion batteries are inexpensive and easy to produce and use. However, there is a known risk of spontaneous combustion, especially in low quality units.¹⁰ The underlying mechanism seems to be a self-induced rapid overheating of the battery.¹⁰ A specific issue of lithium-ion batteries is the known phenomenon called “runaway,” which seems to be responsible for the chain reaction, eventually leading to their explosion. The runaway phenomenon starts with a short-circuit within the battery. The temperature can rapidly rise to 500°C and more, resulting in fire and explosion.¹⁰ According to Palmieri and colleagues, the runaway phenomenon can be triggered by low quality manufacturing, battery overcharging, exposure to high voltage (e.g., due to wrong or failed charger), and exposure to decomposed electrolyte solutions.¹⁰ Furthermore, there are several reports of battery ignition due to contact with metal objects, such as coins or keys in trousers’ pockets.^6,11

The exploding battery usually produces small shrapnel-like particles that can penetrate soft tissue. Moreover, the surrounding clothes can be set afire, leading to serious burn injuries.

Rossheim et al. report data of 2,035 cases of e-cigarette explosions admitted to US emergency departments from 2015 to 2017, with a rising incidence.¹² Colleagues from the United Kingdom recently published a comprehensive review of 90 published cases.⁸

On the basis of available data, it can be summarized that burn injuries associated with e-cigarette explosions are serious issues requiring sophisticated treatment and are leading to relevant morbidity.

Not least because of growing evidence of an associated lung disease, most recently coined Vaping Associated Lung Injury (VALI), which seems to lead to a serious compromise of pulmonary function, health concerns about e-cigarette abuse are still a hot topic.¹³ This is the first study that analyzes not only patient characteristics, injury patterns, and treatment modalities, but also long-term outcome and information on post-injury abuse.

Patients and Methods

This is a single-center retrospective study analyzing data of 46 consecutive patients who presented with injuries after explosion of an ENDS at the Hannover Medical School Burn Center in northern Germany between July 1, 2013 and August 31, 2019. Excluded were all outpatient cases. In addition, patients who suffered burn injuries due to a house fire ignited by ENDS were not considered during this study.

Patient and clinical characteristics, as well as clinical management and outcome, e.g., the onset of wound infection, were captured in an electronic database. In addition to the data from electronic patient files, a standardized telephone interview was performed to collect mid-term results and more specific information regarding vaping habits, treatment results, and patient satisfaction.

After a mean of 13 months post-discharge, 31 patients (67%) were reached and completed the interview. These patients were asked (1) whether they technically manipulated their ENDS before explosion; (2) where they bought their device; (3) whether they continued vaping; and (4) whether they were satisfied with the treatment results.

This study was outlined and performed in full concordance with the Declaration of Helsinki ethical guidelines, and all patients gave their written consent to data storage, analyses, and use of photographs for scientific purposes and publication.

Statistical analysis

In addition to the analysis of descriptive parameters, inferential statistics were performed to identify relevant risk factors for higher injury grades as well as impaired outcome. The Kolmogorov-Smirnov test was performed to assess the distribution of continuous data. If the data was normally distributed, it is shown as mean and standard deviation; if a non-parametric distribution was detected, the data is shown as median and range. For the comparison of non-parametrically distributed data, the Mann-Whitney-U test was applied. In case of parametric distribution, the data were compared with the Student's two-sided t-test. Categorical variables were analyzed with the Pearson's chi-squared test. A p-value <0.05 was defined as significant.

Results

Demographic characteristics

Of 46 included patients, four were females (1.8%); thus, men were significantly more often affected by ENDS-explosion injuries (p < 0.001). The median age of the patients was 36 years (range 18–58 years).

Injury patterns

All included patients presented with at least partial thickness burns. In addition, 39 patients (85%) were affected by deep partial thickness burn injuries. In eight patients (17%), full thickness burns were also observed. In Figure 2, three characteristic injury patterns are shown, that is, burn injuries to the hand (Figs. 2a & b) and the lower thigh (Figs. 2c & d) as well as the upper thigh (Figs. 2 e & f).

Figure 2.

Characteristic injury patterns due to the explosion of an ENDS. a: Injury to the hand – initial appearance. b: Same injury after debridement. c: Injury to the lower thigh – initial appearance. d: Same injury after debridement. e: Injury to the upper thigh – initial appearance. f: Same injury after debridement.

32 patients (69%) had injuries to the waist and groin region. In 12 patients (25%), the hand was affected. A facial injury was observed in 3 patients (7%). In summary, regarding the Patterson classification, there were 25% Type I injuries, 7% Type II injuries, and 69% Type III injuries. No Type IV or Type V injuries were observed (see Table 1). The mean affected total body surface area (TBSA) was 3%.

Table 1.

Classification of ENDS explosion–associated injuries as proposed by Patterson et al. and distribution of the study population.¹⁴

Type	Description	Number of patients (% of study population)
I	Hand injury	12 (25%)
II	Face injury	3 (7%)
III	Waist/groin injury	32 (69%)
IV	Injuries due to house fire ignited by ENDS	Nil
V a	Direct upper airway inhalation injury	Nil
V b	Subglottic inhalation injury	Nil

ENDS, electronic nicotine delivery systems.

Clinical management and outcome

All patients received wound debridement and rigorous disinfection. In nine cases (19%), hydrotherapeutic debridement was indicated. 61% of the included patients (n = 28) were treated with a consecutive operative treatment, that is, tangential necrectomy and subsequent split thickness skin graft transplantation. In 31 patients (67%) a synthetic polylactide skin substitute (Suprathel®, PolyMedics, Denkendorf, Germany) was applied. In all other cases, wounds were treated with the application of polyhexanide wound gel and non-adhering dressings.

Wound infection was observed in 18 cases (39%). Burn depth was the only statistically significant risk factor for the onset of wound infection (p < 0.001). An example of long-term post-operative outcome is shown in Fig. 3.

Figure 3.

Definitive operative treatment and follow-up results (same patient as Fig. 2e & f). a: Definitive surgical treatment with split thickness skin graft transplantation. b: Same patient, follow-up results 24 months after treatment.

Sub-cohort analysis: Vaping habits, ENDS type and acquisition source, treatment satisfaction

Of the entire study population, 31 patients (67%) could be followed up and completed the telephone interview. Of these, 91% (n = 28) were satisfied with the treatment outcome regarding scar appearance. 62% (n = 19) reported that they ceased using ENDS after the injury, and 38% are still using the device. 13 patients (42%) had altered their device either for prolonged battery life or to increase smoke production. 25 patients (79%) bought their ENDS on the internet, whereas six patients (21%) bought their device in a specialized shop.

Discussion

In the literature, a rising incidence of ENDS explosions and resulting burn injuries has been observed since the first reports in 2016.3, 4, 5, 6^,8 The US FDA's center for tobacco products listed 92 confirmed cases of ENDS explosions from 2009 to 2015.¹⁵ On the basis of nationally representative data provided in the United States National Electronic Injury Surveillance System, Corey and colleagues estimated that the numbers of patients treated in US emergency departments are even higher, at more than 1000 cases per year.⁷ Taken together, burn injuries due to exploding ENDS are assumed to be a notable issue with increasing importance in everyday practice.

Therefore, the current study was performed to analyze the patient population of a large burn center with specific focus on the injury patterns, treatment options, and outcome.

Injury patterns and consecutive therapeutic options

Patterson et al. provided a classification of ENDS-explosion-associated injuries in 2016, which divides these into five categories (see Table 1). While injuries to the waist and groin as well as the hand were observed regularly in the current study, the incidence of facial injuries was comparatively low (n = 3; 7% of study population). This might well explain the absence of Type 5 (direct upper airway inhalation) and Type 6 (subglottic inhalation) injuries, which were both associated with explosions near the airway. Injuries due to house fires ignited by ENDS, which are categorized as Type 3 by Patterson and colleagues, were excluded from this analysis.¹⁴

The classification by Serror and colleagues further takes the mechanism of explosion into account. Since this was not documented in the patient files of the current study, application of this classification was not feasible.¹⁶ Serror and colleagues presented data on 10 patients that were treated in their burn center within 1 year. With a mean TBSA of 3%, and 80% thigh as well as 50% hand injuries, they reported comparable injury patterns, with a higher rate of injuries to the hand.¹⁶ They rightfully stated that one of the major issues with these specific injuries is their topography, affecting functionally and cosmetically demanding body areas. Injuries to the hands are mostly due to the fact that the patients try to extinguish a fire from their burning trousers, although they can also due to direct explosions while holding the ENDS. The latter is also the case with facial injuries.

In 2018, Jones and colleagues presented a comprehensive review on a total of 90 patients who suffered injuries from ENDS explosions.⁸ Their cumulative data align with the current results. 95% of the patients included in the review were male and the mean age was 30 years. The mean burned TBSA was 4.9% and the majority of cases had mixed partial and full thickness burns. Contrarily to the current data, Jones et al. reported a smaller number of surgically treated patients of only 39% (n = 35).

This notable disparity might be due to a selection bias, which could be a limiting factor to the present data. More severe cases leading to surgical treatment are more likely to be transferred to a burn center, whereas lesser injuries necessitating only conservative treatment might well be treated in a smaller, peripheral hospital. Another possible limitation of this study is the single-center design. Nevertheless, based on the current results, it should be emphasized that referral to a burn center was warranted in most of the cases. A proper assessment of the injury pattern, including identification of co-injuries, e.g., to the airway or the eyes, and specifically the assessment of burn depth is essential to guarantee a good outcome, not least because burn depth was significantly associated with the onset of wound infection in the present cohort (p < 0.001).

Furthermore, based on the distinctive injury pattern, patients suffering from ENDS-explosion-associated injuries vastly benefit from an interdisciplinary approach including plastic surgeons, emergency physicians, otolaryngologists, and ophthalmologists.

Follow-up results: Patient satisfaction and vaping habits

Only 67% of the study population completed the follow-up interview, although this is considered a comparatively high response rate.¹⁷ Patient satisfaction with treatment outcome regarding scar appearance was high (91%; n = 28); hence, the treatment approach outlined in this study is efficient and can be recommended.

Surprisingly, 38% of patients reported in the follow-up interview that they were still vaping. This counterintuitive and irrational observation might well underline the addiction potential of ENDS-abuse.

It has been discussed that individual alterations of the ENDS are a notable contributor to their explosion and concomitant injuries. In the current cohort, nearly half of the questioned patients (42%) admitted to have modified their devices on their own. Most of them indicated that they did so to prolong battery life or to enhance smoke production for a more satisfying smoking experience. The presented results underline the importance of avoidance of such practices.

Regarding the source of the ENDS, there was no significant difference between purchases from the internet or specialized shops. Therefore, it can be concluded that there is no safer supply source for ENDS.

Public health and prevention initiatives

In the past, it has been argued that ENDS might be a proper tool to foster smoking cessation. However, this is still a matter of controversy, and there is inconsistent evidence.¹⁵ The current understanding is that ENDS may support smoking cessation but only in a small fraction of smokers, based on a US population survey.¹ It should be noted that the dangers of ENDS explosions were not taken into account in the aforementioned studies.

The current results, in addition to literature data, provide good evidence to aim public health initiatives against ENDS-abuse. The presented data supports available evidence that primarily younger men are at risk for ENDS-explosion-associated injuries. Therefore, the major target group of educational preventive campaigns should be males between the ages of 18 and 35.⁸ Another leverage point is the clear association between individual alteration of the device and its explosion as reported by 42% of the analyzed cohort. Individuals should be firmly warned against this practice. There is enough evidence, including this study, which should motivate law makers as well as public health officials to foster strategic prevention programs to elucidate the serious dangers of ENDS. In this context, it should not be disregarded that social media plays an enormous role as a communication channel, especially in the focused group. Interestingly, there is research available that suggests that young adults are commonly exposed to e-cigarette-related marketing or user-generated content on the social media they use.¹⁸

In the near future, the popularity of ENDS might decline due to the recent developments regarding the identification VALI, which leads to an aseptic inflammation of bronchial tissue, acute respiratory distress syndrome, and potential death.^13,19 The US Center for Disease Control declared VALI as a nationwide epidemic.¹⁹

The dread-risk effect

Another interesting aspect to be discussed in this context of this study was analyzed most recently by Nyman and colleagues.¹⁵ In their work on the perceptions of a population of smokers on ENDS-explosion risk, they found out that there is a notable “dread-risk effect.” This means that the perception of a real risk is emotional rather than rational. Hence, an accident, especially when it is very imaginable and dramatic such as an explosion, is much more perceived as a relevant risk than a chronic, invisible thread, such as VALI. Surprisingly so, the rate of patients continuing the abuse is quite high, although 62% of the followed-up patients ceased ENDS-usage.

Nevertheless, the dread-risk effect stresses the importance of research on ENDS explosion risk. It very well can be assumed that the impressive visual dread risk of explosion injuries and the profound evidence on this topic will lead to a significant reduction of ENDS abuse. This might even be the case in comparison to reports on VALI, which appears to be a much more invisible threat to peoples’ health.

Conclusions

Injuries from exploding ENDS can be serious and should be treated in a specialized burn center. ENDS explosions lead to a characteristic injury pattern, affecting the groins and thighs, hands, and face. Patients often need surgical treatment. The serious danger of ENDS explosions should be made more public, on the one hand to address the risk of explosion itself, and on the other hand as a utilization of the “dread-risk” effect, which might lead to overall cessation of ENDS-abuse. It is the responsibility of legislators and medical professionals alike to raise awareness on this issue.

Conflict of Interest

The authors have no competing interest to declare.