Bulletin of the World Health Organization

Oesophageal corrosive injuries in children: a forgotten social and health challenge in developing countries

Sandro Contini a, Alim Swarray-Deen b & Carmelo Scarpignato c

a. Department of Surgical Sciences, University of Parma, Via Gramsci 14, 43100, Parma, Italy.
b. Emergency Surgical Center, Goderich, Sierra Leone.
c. Department of Anatomy, Pharmacology and Forensic Sciences, University of Parma, Parma, Italy.

Correspondence to Sandro Contini (e-mail: sandro.contini@unipr.it).

(Submitted: 16 September 2008 – Revised version received: 06 December 2009 – Accepted: 27 February 2009.)

Bulletin of the World Health Organization 2009;87:950-954. doi: 10.2471/BLT.08.058065


An unsafe environment is a substantial risk factor for child injury and violence, therefore representing a significant cause of child death and disability especially in developing countries, where 95% of all child injury deaths occur.1,2 Among those injuries that are caused by an unsafe environment, the accidental ingestion of corrosive substances is declining in high-income countries35 but not in developing countries,6 where it is quite significant, especially among illiterate people with poor socioeconomic status. Nevertheless, information about prevention, management and outcome of these common accidents in children of low- and lower-middle income countries is scarce.

The Italian nongovernmental organization, Emergency, is currently running five surgical centres in Afghanistan,3 Cambodia1 and Sierra Leone,1 mainly for the surgical treatment of war and civilian trauma injuries. In the first years of Emergency’s work in Sierra Leone, which began in November 2001, an unexpectedly high number of children were admitted after accidental caustic injuries, most of them having been treated at home or by local doctors, often resulting in oesophageal perforation and death. Between December 2005 and June 2008, 148 children were admitted to the hospital for accidental caustic soda ingestion. To meet this increasing medical need, at the end of 2005 the hospital was supplied with paediatric and adult upper gastrointestinal fibre-optic endoscopes and with dilatation devices (Savary bougies, balloon catheters) to manage these injuries either in the acute or in the late phase when severe oesophageal strictures can occur.7 Stimulated by this high number of patients, a medical literature review was performed from 1990 to 2007 (inclusive) for papers about caustic ingestion coming from low- or lower-middle income countries.8 The data sources included four independent databases: MEDLINE, EMBASE, SciELO and LILACS.

Situation analysis

Only 37 papers were found concerning corrosive ingestion in children of low- and lower-middle income countries. Ninety-four per cent were published by referral hospitals and 59% came from Africa. Only eight papers concerned the specific epidemiology of caustic ingestion. Children aged less than 5 years were most frequently injured (80%), and boys exceeded girls (70%). The most reported corrosive agent was caustic soda, followed by kerosene, sodium hypochlorite and generic household chemicals. The ingestion of acid was more common in India than in other countries. Corrosive ingestion accounted for 0.3% of paediatric admissions in the Gambia9 and for 0.5% in Nigeria,10 thus amounting to a 0.84% of total childhood mortality in that country. Death rates ranged from 0 to 11.9% (mean 4.1%) although the rate of late oesophageal strictures can reach 50% in the presence of early severe oesophageal lesions.11,12

This type of injury is largely unreported in developing countries and its true prevalence simply cannot be extrapolated from random articles or personal experience. The data available are heavily skewed towards well-resourced centres and do not reflect the entire reality of the condition. District hospitals probably observe most of these children, but their caseload is largely unknown. Sometimes children may not arrive at the hospital, either because they are too ill and die, they live too far away or their lesions are not severe. Occasionally families cannot afford the cost of long-term and complex therapies. With such a lack of data, it is hard to estimate, even roughly, the proportion of patients who do not seek medical care from hospitals.

Most ingestions are due to parent’s lack of knowledge of the hazards of corrosive substances kept in the house, crowded living conditions in slums and the availability of chemicals in and around the houses, combined with the natural curiosity of children. The widespread lack of any preventive measures is the strongest risk factor for these injuries.


All children admitted for caustic ingestion before December 2005, and still being followed up, were recalled to assess them endoscopically and to submit them to a dilatation, if needed. While the programme was aimed at treating children coming from areas nearby, a progressive increase in admissions was observed as patients were referred by international nongovernmental organizations from distant towns or villages.

Patients admitted soon after injury (48–72 hours after ingestion), and with absent or mild oesophageal lesions at endoscopy, were usually discharged. Children with more severe oesophageal findings or severe dysphagia were admitted and a second endoscopic examination/dilatation, was scheduled after 3 weeks. The most significant therapeutic option in the acute post-injury phase was a surgical gastrostomy, performed in children unable to swallow liquids or saliva, to achieve adequate nutritional support. When patients arrived at the hospital after a delay of 72 hours to 3 weeks after ingestion, endoscopy was not carried out due to the high risk of perforation. A gastrostomy was performed in the presence of severe dysphagia or after an unsuccessful dilatation attempt in children admitted late, i.e. at more than 3 weeks after the injury, with swallowing problems. The first dilatation was always carried out at least 3 weeks after ingestion. This management strategy is represented in Fig. 1.

Fig. 1. Flowchart for the management of corrosive ingestions in children, adopted at the Emergency Surgical Center in Goderich, Sierra Leone
Fig. 1. Flowchart for the management of corrosive ingestions in children, adopted at the Emergency Surgical Center in Goderich, Sierra Leone

Dilatations were performed at 7–10 day intervals, usually by means of rigid Savary bougies (93.6%), while balloon dilatation was employed only in a few patients (9.6%) that we treated at the beginning of our research project. Guide wires and dilators were preferably introduced through the gastrostomy (if performed) by a retrograde approach. The gastrostomy was removed after a period of at least 8–12 months without dysphagia. Mandatory reference points for a successful outcome were considered to be: a long-lasting oesophageal patency together with an improvement in nutritional status, assessed by weight-for-height parameters13 (if < 80%, an increase to > 80%; if > 80, increase of 1 standard deviation). A training programme for gastrostomy and post-dilatation feeding was carried out with the patients’ mothers to help meet these requirements.


From December 2005 to July 2008, 148 children (aged 14 months to 15 years; mean 4.5 years; 58.4% males) were admitted for accidental caustic ingestion. Only 29 of these (19.5%) were admitted early; two of them (6.8%) with severe respiratory tract damage leading to death. Twenty showed mild or no lesions and were discharged. All other children (119) were admitted several days, weeks, even months after ingestion, complaining of severe dysphagia. Overall, 126 children were submitted to dilatation, with a mean of 4.9 (range: 1–23) procedures per child. A gastrostomy was done on 92 of 126 children (i.e. 73%).

Recurrent strictures were experienced in 54 of 126 (42.8%), and were significantly more frequent in late oesophageal narrowing.14 Twenty-four (19%) children are still on a continuous dilatation programme. In 3 patients it was not possible to overcome the stricture and two of them were sent for oesophageal replacement elsewhere. Perforation was observed in seven patients (5.6%) with two deaths, both after balloon dilatation. Successful (as defined previously) dilatations were obtained in 96 children (i.e. 76%). The total death rate was 4% (5/126).


Corrosive ingestions by children in developing countries have some peculiar features (Box 1). The most frequently ingested substances, such as caustic soda, have a powerful solvent action that results in very serious injuries, as confirmed by 73% of children needing gastrostomy feeding, in sharp contrast with the 11% of severe injuries reported in a multicentric study from a high-income country.15 Many injuries do recur, with one-fifth of patients still requiring a continuous dilatation programme. Unfortunately, long-term dilatation programmes are very challenging in low-income countries. Repeated hospital visits cost money, may result in loss of work for parents and neglect of other children at home. After the first few visits to the hospital, the parents may get exhausted and start feeling frustrated. These socioeconomic and psychological factors must then be taken into account in the long-term follow up.

Box 1. Lessons learned

The majority of oesophageal caustic strictures in children are observed late, when dilatation procedures are likely to be more difficult and carry a significantly higher recurrence rate.

Gastrostomy is necessary to maintain adequate nutritional status but mothers need training to feed their children this way.

Both improvement in nutritional status and sustained oesophageal patency should be the reference points to a successful dilatation.

Patients are frequently treated at home by traditional therapies, by witch doctors or by physicians working in district hospitals without specific expertise in the field. They may require several days of travel to reach the hospital, meaning timely evaluation and treatment of these accidents is unlikely. Furthermore, frequently children are admitted when the stricture is already well established. Late dilatations are more difficult and followed by a significantly higher recurrence rate than early procedures.13 A delayed presentation and treatment has been considered a strong predictor for a future oesophageal replacement.16

Gastrostomy is necessary to feed patients and to keep them alive. However, feeding through gastrostomy or after dilatation may be followed by progressive malnutrition due to the inability of the families to nourish children properly and the lack of appropriate feeding solutions. Mothers should be trained and helped in feeding techniques. Moreover, gastrostomy is useful for a retrograde dilatation, as it is less risky and also provides the option to leave a string running through the stricture, from the gastrostomy through the nose. This is particularly helpful in cases of severe stricture that are difficult to overcome using guide wires.

When evaluating the outcome of treatment, both an improvement in nutritional status and sustained oesophageal patency, with an adequate lumen to guarantee normal food intake for growth and development, should be the reference points to a successful dilatation. Defining risk factors and groups at risk, providing appropriate education and enforcing regulations for manufacturers of household products would certainly help to reduce significantly the number of fatalities. Governments should play their role in educating people, but their efforts (if any) may be unsuccessful due to the high rate of illiteracy and to poorly resourced health systems. Even humanitarian organizations could make efforts in this direction but, again, there has not yet been enough focus on the seriousness of the problem. ■


This work has been possible thanks to the skilful cooperation of the staff of the Italian nongovernmental organization Emergency either at the Emergency Surgical Center in Goderich (Sierra Leone) or at the Milan (Italy) HQ.

Funding: The programme of treatment of corrosive injuries at the Emergency Surgical Center in Sierra Leone was financially supported by the Fondazione della Cassa di Risparmio di Parma, Parma, Italy.

Competing interests: None declared.