Acid and Alkali Chemical Assault Burns (Vitriolage): A Systematic Review of Epidemiology

Abstract

Background:

Chemical assaults involve the intentional use of chemicals to cause harm. These attacks frequently occur in domestic settings, with reliable epidemiological data remaining limited. This systematic review sought to explore epidemiological aspects of such incidents.

Methods:

It included all peer-reviewed reports providing original cases of chemical assaults, detailing at least the number of victims involved, identified through PubMed and Web of Science searches conducted from database inception to May 26, 2024.

Results:

Globally, the mean age of victims was 29.03 years (SD 11.99 y). Women accounted for 50.71% of the victims, whereas men were 89.01% of the perpetrators. Over half of the victims were illiterate (53.76%), and homemakers represented the most affected occupation (33.84%). Assaults most frequently occurred on the street (47.40%). Acids were the most used substance (70.36%). Victims were more likely to have a preexisting relationship with their perpetrator (58.56%). The most common motive for chemical assaults was land/property disputes (16.72%). The head, face, and eyes accounted for 39.01% of all body parts burned. The mean total body surface area burned was 14.12% (SD 15.42%).

Conclusions:

Although the results do provide insights, any interpretation must be approached with caution, as several variables are supported by only sparse data. Future recording of chemical assaults requires a more formal and coordinated approach, ideally through databases that involve hospitals, law enforcement, and government agencies.

Takeaways

Question: What are the epidemiological characteristics of chemical assaults (vitriolage) reported in the literature?

Findings: This systematic review revealed that globally, the mean age of victims was 29.03 years. Women accounted for 50.71% of the victims, whereas men were 89.01% of the perpetrators. Most variables were supported by only sparse data, meaning any interpretation must be approached with caution.

Meaning: There is a critical need to improve data recording practices, ideally through the establishment of collaborative databases that integrate input from hospitals, law enforcement, and governmental agencies.

INTRODUCTION

Unlike large-scale chemical attacks associated with terrorism or warfare, domestic chemical assaults are often motivated by interpersonal conflicts.¹ These assaults, often premeditated and involving strong acids,² can have devastating effects on the physical and psychological health of victims³ and pose significant challenges for medical and law enforcement professionals. Chemical assaults first became prevalent during Victorian times, largely due to the widespread availability of sulfuric acid, then known as “vitriol,” that accompanied the Industrial Revolution.⁴ This led to the coining of the term “vitriolage,” defined as “an intentional act of violence in which any corrosive is thrown onto the face and body of a person to disfigure them.”⁵ These assaults are now a global issue, gaining significant media attention, including the 2012 Pakistani documentary “Saving Face” and the 2020 Indian film “Chhapaak” (meaning “splash” in Hindi).⁶ Often, the face of the victim is targeted, aiming to inflict maximum visible disfigurement, as opposed to homicidal intent.⁶

In an attempt to reduce the frequency of chemical attacks, governments have introduced specific criminal acts, including “The Natalia Ponce Law 2015,” in Colombia, which increased maximum sentences for those found to be guilty of chemical assaults,⁷ and “The Acid Law 2012,” in Cambodia, which regulated the production and distribution of concentrated acids.⁸ However, due to unreliable reporting of incidents, it remains unclear if a reduction in chemical assaults has occurred.⁹ Nongovernmental organizations play a significant role in aiding victims, particularly in developing countries, where access to financial and legal support, as well as medical treatment, is often limited. Recognizing these challenges, the UK-based nongovernmental organization Acid Survivors Trust International has collaborated with the United Nations, various law firms, and the British government.¹⁰ Although the legal aspects of chemical assaults have been widely discussed in the literature,^1,5,11 epidemiological data remain notably scarce, existing mainly in the form of isolated case series. This review aimed to look at key epidemiological aspects of chemical assaults, including the incidence across regions, demographic trends, motives, and the types of substances used.

METHODS

General Information

This study was not prospectively registered. The authors acknowledge this lapse and confirm that the study adhered to the planned methodology and objectives, in accordance with the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines.

Eligibility Criteria

Our study included original materials, including conference abstracts, and excluded review articles. Eligible reports provided original cases of chemical assaults, detailing at least the number of victims. We focused on interpersonal (domestic) assaults, excluding terrorism, war-related incidents, self-harm, and assaults involving hot water, flame, or grease. No restrictions were applied to publication year, setting, or language. Chemical assaults were broadly defined as purposeful exposure to a corrosive substance.

Information Sources

Relevant studies were identified through PubMed and Web of Science searches conducted from the inception of each database through May 26, 2024. No further database searches were conducted thereafter. Additionally, a snowball search was performed by reviewing the reference lists of eligible publications.

Search Strategy

The terms (vitriol*[Title]) OR ((acid[Title] OR acidic[Title] OR alkali[Title] OR chemical[Title] OR corrosive[Title] OR bleach[Title] OR lye[Title]) AND (attack[Title] OR assault[Title] OR weapon[Title] OR violence[Title] OR throw[Title] OR crime[Title])) were used.

Selection Process

Results were exported to Covidence, where abstracts were screened for relevance, followed by full-text reviews. Two reviewers independently assessed the studies at each step, with discrepancies resolved by a third reviewer. Google Translate was used for translations, and no automation tools were used in the selection process. Authors were contacted directly when full-text access was unavailable.

Data Collection Process

A data extraction table was designed and completed by 2 reviewers in a blinded fashion. Extracted data were compared, with discrepancies resolved by a third reviewer. No automation tools were used. When the data were unclear, authors were contacted for clarification.

Data Items

In the included reports, the primary outcomes extracted encompassed several key domains. Victim demographics included the number of victims, sex, age, education level, and occupation. The only perpetrator characteristic documented was sex. Assault features were recorded, including country, specific location, time of day, causative agent, victim–perpetrator relationship, and motivation. Burn characteristics were described in terms of the percentage of total body surface area (TBSA) affected and the anatomical distribution of injuries. No secondary outcomes were reported.

Assault timing was categorized as “day” or “night,” with further distinctions of “morning,” “afternoon,” and “evening” for daytime incidents. If the time was unspecified, it was classified as “not reported.” Chemical assault motivations were classified by “romantic,” “nonromantic,” or “unclassified.” “Romantic” motivations were defined as those related to romantic relationships, or sexual matters, whereas “nonromantic” motivations involved no such connections. We also recorded the study design and year of publication.

Study Risk of Bias Assessment

To assess study quality and risk of bias, we used the Joanna Briggs Institute Critical Appraisal Tools for case series and reports, evaluating methodological quality and potential biases. The full tools are available at https://jbi.global/critical-appraisal-tools (accessed May 14, 2024).

Effect Measures

For continuous variables, we calculated the mean and SD, with no statistical analysis performed on categorical variables. A meta-analysis was not conducted, as the review focused on the epidemiology of chemical assaults rather than health interventions.

Synthesis Methods

Continuous data were synthesized only from studies providing a mean or sufficient data for calculation. No restrictions were applied to categorical data. For case series with individual reports, means and SDs were calculated across all variables where possible. The global distribution of victims was shown on a choropleth map, causative agents with a pie chart, and affected body parts using an anatomical diagram. Other data were presented in tables or text. Structured summaries were used to concisely overview the data for each synthesis. A follow-up paper will address regional heterogeneity, and no sensitivity analysis was performed due to the observational nature and variability of the studies.

Reporting Bias Assessment

Using the Joanna Briggs Institute Critical Appraisal Tools, case series were evaluated across 10 domains and case reports across 8. Scores for case series were categorized as low (1–4), moderate (5–7), and high (8–10), and for case reports, as low (1–4), moderate (5–7), and high (8). All evaluations were performed by 2 independent reviewers, with disagreements resolved by a third. No automation tools were used. Missing, unclear, or inapplicable data received a score of zero.

Certainty Assessment

A certainty analysis was not performed due to the lack of standardized interventions, control variables, and the observational nature of the studies.

RESULTS

Study Selection

Our literature search yielded 3227 results, with 238 identified as duplicates. After screening, 2773 articles were excluded for not being specifically related to chemical assaults. The remaining 210 papers were assessed for eligibility. Of these, 45 were excluded for focusing on chemical assaults in acts of war, 28 for not presenting original cases, 28 for lacking relevance to chemical assaults, 10 due to the unavailability of the text, and 8 because they were review articles. The flow of studies through the review process is illustrated in Figure 1 using a modified PRISMA flowchart.

Fig. 1.

PRISMA flowchart illustrating the review’s identification, screening, eligibility, and inclusion stages.

Study Characteristics

The 90 eligible studies included 3599 chemical assault victims, consisting of 74 case series, 18 case reports, and 2 studies with both. Supplemental Digital Content 1 summarizes demographic data and burn characteristics.^2,11–99 (See table, Supplemental Digital Content 1, which displays a table summarizing the demographics and burn characteristics of chemical assault cases. Studies focusing on ocular injuries were marked as N/A for TBSA, https://links.lww.com/PRSGO/E552.)

Risk of Bias in Studies

Two studies presented case series and case reports as separate entities. Of the 74 case series, 49 were high quality (66.2%), 15 were medium (20.3%), and 10 were low (13.5%). Of the 18 case reports, 7 were high quality (38.9%), 9 were medium (50%), and 2 were low (11.1%). Quality and risk of bias appraisals are provided in Supplemental Digital Content 2. (See table, Supplemental Digital Content 2, which displays quality assessments and risk of bias evaluations for all included studies, https://links.lww.com/PRSGO/E553.)

Results of Syntheses

Risk of bias analysis for individual syntheses is available in Supplemental Digital Content 3 and 4. (See table, Supplemental Digital Content 3, which displays the risk of bias analysis for each synthesis included in the review, https://links.lww.com/PRSGO/E554.) (See table, Supplemental Digital Content 4, which presents the risk of bias analysis for each study included in the review, https://links.lww.com/PRSGO/E555.)

Victim Demographics

The 90 studies included a total of 3599 victims of chemical assaults. Of these, 1905 victims (52.93%) had their sex recorded, showing a slight female majority: 966 women (50.71%) and 939 men (49.29%), resulting in a male-to-female ratio of 0.97:1 (female-to-male ratio of 1.03:1). Age data were provided by 39 studies, representing 537 victims (14.92%), with a mean age of 29.03 (SD 11.99) years. Education level data were available in 5 studies (359 victims, 9.97% of the cohort), revealing 193 illiterate victims (53.76%), 141 with primary or secondary education (39.28%), and 25 with tertiary education (6.96%). Occupational data from 11 studies (331 victims, 9.20%) are presented in Table 1.

Table 1.

Distribution of Chemical Assault Victims by Occupation

Occupation	No. Reported Chemical Assault Victims, %
Unknown/not reported	3268 (90.80 of victims)
Domestic/homemaking	112 (33.84)
Unemployed	57 (17.22)
Student	43 (12.99)
Agriculture	38 (11.48)
Finance and business	32 (9.67)
Laborer	20 (6.04)
Sales and hospitality	15 (4.53)
Transportation	5 (1.51)
Education	2 (0.60)
Engineering and technology	2 (0.60)
Healthcare	1 (0.30)
Garment worker	1 (0.30)
Security	1 (0.30)
Model	1 (0.30)
Journalist	1 (0.30)

Perpetrator Demographics

Twenty-two studies (191 perpetrators, 5.31% of the cohort) provided data on perpetrator sex. Men were the majority, comprising 170 (89.01%) compared with 21 women (10.99%), resulting in a male-to-female ratio of 8.10:1 (female-to-male ratio of 1:0.12).

Location of Assaults

Across the 90 studies, 32 countries reported original chemical assault cases. Figure 2 displays the number of cases by country. Location data were available for 327 cases (9.1%), with the majority occurring on the street (47.40%), followed by at home (42.81%). Other locations included public transportation (3.06%), entertainment venues (2.14%), workplaces (2.14%), shops (0.61%), cars (0.61%), restaurants (0.31%), schools (0.31%), hostels (0.31%), and banks (0.31%).

Fig. 2.

Global distribution of chemical assault victims by country.

Time of Assaults

Three studies (197 victims, 5.47% of the cohort) provided data on the timing of chemical assaults. Daytime assaults were more common (56.35%) than nighttime assaults (43.65%). Of daytime assaults, 66.67% occurred in the morning, 15.32% in the afternoon, and 18.02% in the evening.

Causative Agents

Figure 3 shows the distribution of chemical assault victims by causative agent. The majority (62.91%) were caused by unspecified acids. A specific chemical was provided in 478 cases (13.28% of the cohort).

Fig. 3.

Graphical distribution of reported chemical assault victims by causative agent.

Motivation and Victim–Perpetrator Relationship

Table 2 presents the victim–perpetrator relationship in chemical assault cases. Of the 222 cases (6.17% of the cohort) with reported relationship data, 58.56% (130 cases) involved a connection, whereas 41.44% (92 cases) did not. A large number of cases (3377) did not specify whether a prior relationship existed.

Table 2.

Distribution of Chemical Assault Victims by Victim–Perpetrator Relationship

Relationship Between Victim and Perpetrator	No. Reported Chemical Assault Victims
Unknown/not reported	3377 (93.83% of the cohort)
Relationship present	130 (58.56%)
Not specified	57
Romantic partner	43
Spurred lover	7
Ex-romantic partner	3
Sibling	2
Sibling-in-law	2
Uncle/aunt	2
Mistress	2
Neighbor	2
Parent	2
Family of spurred lover	2
Colleague	1
Cousin	1
Psychiatric patient	1
Landlord	1
Stepparent	1
Self-inflicted*	1
No relationship present	92 (41.44%)

^*Included, as it accounted for 1.7% of cases in a relevant study.

Table 3 presents the motivations behind chemical assaults, with land/property disputes being the most common (16.72%, 153 victims). A significant number of cases (2684) had unknown or unreported motivations.

Table 3.

Distribution of Reported Chemical Assault Victims by Motivation of Perpetrator

Motivation of Perpetrator	No. Reported Chemical Assault Victims, %
Unknown/not reported	2684 (79.58 of the cohort)
Nonromantic	404 (44.15)
Land/property dispute	153 (16.72)
Miscellaneous	129 (14.10)
Family dispute	80 (8.74)
Hate crime	24 (2.62)
Robbery/theft	18 (1.97)
Romantic	317 (34.64)
Denial/refusal of love or sex	121 (13.22)
Affair/suspected affair	91 (9.95)
Miscellaneous	67 (7.32)
Dowry	38 (4.15)
Unclassified	194 (21.20)
Accident*	61 (6.67)
Wrongly victimized	60 (6.56)
Collateral damage/bystander	38 (4.15)
Miscellaneous	35 (3.83)

^*Included, as further information suggestive of assault often emerged when the patient felt confident enough to disclose it.

Burn Characteristics

Thirteen studies, covering 291 cases (8.09% of the cohort), provided a mean victim TBSA and SD, or sufficient data for calculation, yielding a combined mean of 14.12 (SD 15.42). Figure 4 presents data on the frequency of affected body parts in chemical assaults. The face, head, and eyes were the most commonly targeted, being affected a total of 928 times (39.01% of the 2379 body parts reported).

Fig. 4.

Distribution of chemical assaults by affected body parts.

DISCUSSION

Although the results offer valuable insights, caution is necessary when interpreting them, as many variables have limited data relative to the overall dataset, potentially reducing the representativeness of the findings. Of all variables, victim sex was recorded in the highest percentage of cases (52.93%), whereas the victim–perpetrator relationship (6.17%), time of day (5.47%), and perpetrator sex (5.31%) were among the least reported. Furthermore, publication bias remains a concern, possibly leading to overrepresentation of high-profile cases, potentially skewing the overall understanding of chemical assaults.

Chemical attacks are often viewed only as acts of sex-based violence, perpetrated by men against women to coerce or control, often in the context of a romantic relationship. Although this is often the case, our findings show that not all chemical assaults are rooted in gender-based violence or romantic relationships.¹ Criminal activities and personal conflicts were found to be common motivators, with some incidents occurring in clusters, such as during consecutive years at London’s Notting Hill Carnival.¹⁹ Despite very limited data, over half of the victims with a recorded education level were classified as illiterate, indicating a potential correlation with lower socioeconomic status.¹⁰⁰ Low literacy is known to create barriers to employment and increase dependency on others for basic needs.¹⁰¹ Economic vulnerability, which may also help explain the high incidence of chemical assaults against homemakers, can also reduce access to legal and social protections. Furthermore, illiteracy limits awareness of rights and available support services,¹⁰² contributing to a greater social isolation and heightened susceptibility to violence.¹⁰³

Despite the insights gained, several limitations must be considered. The reliability of data is often compromised by widespread underreporting of chemical assaults, as victims may fear retaliation, leading to an underrepresentation of the true scale and severity of the issue.^8,9 Additionally, media-reported cases, frequently cited in the literature, may lack accurate injury details, especially when victims avoid medical care.¹⁰⁴ Even when victims do seek treatment, injuries are often misreported as accidental.⁸ Furthermore, many studies are based on small sample sizes or isolated case reports, lacking crucial data on assault characteristics, burn injuries, and victim demographics. Finally, the lack of certainty regarding the accuracy of using Google Translate in analyzing medical data may present challenges around misinterpretation.

In summary, these findings underscore the critical need for improvements in data recording practices, thereby enabling healthcare providers and government agencies to effectively target their interventions toward the most affected demographic groups. Moving forward, the recording of chemical assaults requires a more formal and coordinated approach, ideally through collaborative databases involving hospitals, law enforcement, and governmental agencies. To support this, relevant authorities should consider implementing mandatory reporting mechanisms, similar to the existing requirements for recording gunshot and knife-related injuries in the United Kingdom.¹⁰⁵ Finally, in addition to in-depth studies assessing the effectiveness of current laws in preventing attacks, intercountry analyses are likely to provide valuable insights, driving the development of more nuanced, region-specific approaches.

DISCLOSURE

The authors have no financial interest to declare in relation to the content of this article.

ACKNOWLEDGMENT

This article builds upon the authors’ earlier findings presented in abstract form. During the preparation of this work, the authors used OpenAI’s ChatGPT (GPT-4, Version September 2023) to check for grammatical errors and improve readability. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.