Sie sind auf Seite 1von 6

Pediatr Surg Int (2004) 20: 824–828 DOI 10.1007/s00383-004-1294-4

ORIGINAL ARTICLE

ORIGINAL ARTICLE

Didem Baskın Æ Nafiye Urgancı Æ Latif Abbasog˘lu Canan Alkım Æ Mehmet Yalc¸ın Æ C¸ etin Karadag˘ Nihat Sever

A standardised protocol for the acute management of corrosive ingestion in children

Accepted: 24 August 2004 / Published online: 9 November 2004 Springer-Verlag 2004

Abstract Oesophageal strictures developing after caustic ingestion in children are a serious problem, and several protocols to prevent stricture formation have been proposed. A prospective clinical trial was conducted for preventing strictures in caustic oesophageal burns in a single clinic, and the results are presented. All children with caustic ingestion who had oesophagoscopy for diagnosing the severity of the burn were included in the study. Eighty-one children were included in the series, with ages ranging between 3 months and 12 years. The patients were given nothing by mouth until oesopha- goscopy. IV fluids, broad-spectrum antibiotics, raniti- dine, and a single-dose steroid were given. Oral burns were positive in 66 patients. Oesophagoscopy revealed a normal oesophagus in nine patients, grade 1 burn in 24, grade 2a in 21, grade 2b in 23, grade 3a in two, and grade 3b in one. Patients with grade 1 and 2a burns were discharged after oesophagoscopy. Patients with grade 2b and all grade 3 burns were given nothing by mouth for a week except water when swallowing their saliva, and were fed via total parenteral nutrition. After the 1st week, if there was no problem with swallowing, liquid foods were introduced. No intraluminal tubes were used. At the end of the 3rd week, a barium meal was administered and an upper gastrointestinal series

D. Baskı n (& )

Pehlivanyan ı Sok. 7/4, 34934

_

Mecidiyeko¨y, Istanbul, Turkey

E-mail: didibas@tnn.net

D. Baskı n Æ L. Abbasog˘lu Æ M. Yalc¸ ı n Æ C¸ . Karadag˘ Æ N. Sever

Department of Pediatric Surgery,

S¸is¸li Etfal Education and Research Hospital,

_

Istanbul, Turkey

N. Urgancı

Department of Pediatric Gastroenterology,

S¸is¸li Etfal Education and Research Hospital,

_

Istanbul, Turkey

C. Alk ım

Department of Adult Gastroenterology,

S¸is¸li Etfal Education and Research Hospital,

_

Istanbul, Turkey

taken. Dilatation was performed at 2-week intervals for strictures, which developed in one grade 2a patient, six grade 2b patients, and the grade 3b patient. Only one of these patients is currently on an oesophageal dilatation program. Limiting oral intake and avoiding foreign bodies in the oesophagus seem to provide a good success rate; however, further prospective studies are needed to decrease the incidence of corrosive oesophageal stric- tures.

Keywords Oesophagus Æ Corrosive ingestion Æ Caustic burns Æ Stricture

Introduction

After authorities heeded advice to require standardisa- tion of the dilutions of corrosive agents, cases of cor- rosive oesophagitis decreased in Turkey during the 1980s. Subsequently, as new chemical agents for domestic cleaning were introduced through formal market channels, they also began to find their way onto the informal market, being sold in any kind of empty bottle or paper bag at a lower price. Alkali agents are mainly kept in used water bottles and have been mis- taken for water, and acid agents look like candy and have been swallowed. As a result, within the last decade the number of cases of corrosive oesophagitis has begun to increase once more. Frequently, no information is available about the concentration of the agents involved because of the absence of labelling on containers. The victims are usually the children of families of low edu- cational level and economic status. Oesophageal burns may result in stricture formation. There is no standard treatment of caustic ingestion during the acute phase; several treatment strategies are described in paediatric surgery books [2, 11]. In our prospective study of corrosive oesophageal burns, treatment was planned to decrease the effects of infec- tion on injured oesophageal wall tissue, thus decreasing the incidence of stricture formation.

Patients and methods

Between 1988 and 2003, 168 children were hospitalised for corrosive ingestion. The oesophagoscope was intro- duced to our department in 1997, and subsequently, oesophagoscopy was routinely performed in all cases of corrosive ingestion. Patients treated between 1997 and 2003 were included in the study.

When admitted to the ward, the patients were given nothing by mouth until oesophagoscopy. IV fluids and broad-spectrum antibiotics with a single-dose steroid to decrease the oedema were given. IV ranitidine was added to the treatment to decrease the effects of possible gas- tro-oesophageal reflux. Oesophagoscopy was performed within 24–48 h, ei- ther with sedation in gastroenterology units using a flexible endoscope or under general anaesthesia in the theatre using a rigid oesophagoscope. Oesophagoscopy on patients with severe burns in their oral mucosa was delayed until sufficient healing had occurred. Endo- scopic grading was carried out as previously described [11]; see Table 1. Patients with grades 1 and 2a burns were discharged

if

no treatment was required for their oral burns and/or

if

they swallowed well. An oral anaerobic antibiotic and

a

proton pump inhibitor were given for 3 weeks for

grade 2a burns. These patients were advised to have soft foods for 3 weeks and to drink a glass of water after every meal (Table 2). Hospitalisation was continued for the patients with grades 2b and 3 burns. They were given nothing by

Table 1 Endoscopic grading of injury severity [11]

Grade

Appearance

Grade 0

Normal Oedema and hyperaemia of mucosa Friability; haemorrhage; erosion blisters, exudates or whitish membranes; superficial ulcers Grade 2a plus deep discrete or circumferential ulceration Small scattered areas of necrosis, areas of brownish-black or grey discoloration Extensive necrosis

Grade 1

Grade 2a

Grade 2b

Grade 3a

Grade 3b

825

mouth for a week except water when swallowing their saliva, and total parenteral nutrition (TPN) was added to their treatment. After the 1st week, if there was no problem with swallowing, liquid foods were introduced, and only soft foods were advised until the end of week 3. When they were discharged, an oral anaerobic antibiotic and a proton pump inhibitor were given for another 2 weeks (Table 2), again with advice to drink a glass of water after every meal. At the end of the 3rd week in all patients except the ones with grade 1 burns, a barium meal was adminis- tered and an upper gastrointestinal (GI) series taken to investigate the oesophageal passage, stricture formation, gastric emptying, and gastro-oesophageal reflux. If stricture formation was seen, dilatation was per- formed every 2 weeks beginning the 4th week after ingestion. Dilatation was carried out either with Savary- Gilliard dilators by means of a flexible oesophagoscope under sedation, or with Rusch dilators or balloon dila- tors under general anaesthesia. In cases of oesophageal perforation or intractable strictures interfering with oral feeding, a feeding gastrostomy was opened. Samples of oesophageal tissue were taken for biopsy in patients requiring frequent dilatations.

Results

Of 118 patients treated between 1997 and 2003, a total of 37 were excluded from the study. Of these, 13 patients presented for treatment after formation of a stricture; the time of ingestion was not noted in three patients; and 21 patients either did not have oesophagoscopy, the examination note for the oesophagoscopy was missing, or the patient’s parents refused permission for an oesophagoscopy. Eighty-one children were included in the series, 37 girls and 44 boys. The mean age was 4 years (range 3 months–12 years). Our youngest patient had swal- lowed a caustic agent given by his mother in place of an antipyretic by mistake. Seventy patients presented within 24 h, eight within 48 h, and three within 1 week of ingestion of the corrosive substance. Thirty-seven patients had acid ingestion, 32 had alkali ingestion, and

Table 2 Treatment protocol for acute caustic ingestions after evaluation with oesophagoscopy (TPN total parenteral nutrition)

Grade

Steroid

Antibiotic a

H2 receptor

Oral feeding a

TPN a

Hospitalisation

Follow-up

 

blocker

 

duration a

1

Single dose

Single dose

No

Yes

No

Discharge after

Not needed

 

oesophagoscopy

2a

Single dose

3 weeks, single anaerobic 1 week broad-spectrum, 2 weeks single anaerobic 3 weeks broad-spectrum

3

weeks

Yes

No

Discharge after

Barium meal after 3 weeks Barium meal after 3 weeks Barium meal after 3 weeks

   

oesophagoscopy

2b

Single dose

3

weeks

Nil for 1 week except water Nil for 1 week, except water

1 week

1 week

3a and 3b

Single dose

3 weeks

1 week

1 week

826

Table 3 Comparison of oral and oesophageal burns

Oesophageal

No

Oral

Oral

Oral

Total

grade

oral burn

grade 1

grade 2

grade 3

0

4

5009

1

3

3

18

0

24

2a

4

14

2

1

21

2b

2

12

9

0

23

3a

2

0

0

0

2

3b

0

0

0

1

1

Total

15

34

29

2

80 a

a Esophagoscopy could not be performed in one oral grade 3 pa- tient

12 had ingestion of other corrosives. All ingestions were

accidental. On admission, 59 patients had difficulty swallowing. Perioral skin and/or oral mucosal burns of varying de-

grees were present in 66 patients (Table 3). One patient had blood-stained vomiting, and two had abdominal pain. Oesophagoscopy revealed a normal oesophagus in nine patients. Grade 1 oesophageal burns were seen in

24 patients. Twenty-one patients had grade 2a oesoph-

ageal burns, 23 had grade 2b, two had grade 3a, and one had grade 3b. Of these patients, four patients with grade 2b burns, two with grade 2a, and one with grade 1 had erosive gastritis (Table 4). In one patient,

oesophagoscopy could not be performed because of se- vere grade 3 oral burns. Patients with no burn on the oesophagus or with grade 1 burn were discharged after 1–6 days (mean 4 days), depending on their oral burn treatment. One of them received short-term TPN because oral burns pre- vented oral feeding. None of these patients had devel- oped an oesophageal stricture (Table 4). Five of the 21 patients with grade 2a burns received TPN. They were discharged after 1–11 days (mean 5.5 days). One of them developed a stricture 3 weeks after the burn, and gastro-oesophageal reflux was pres- ent on her upper GI barium meal. Another patient showed no signs of swallowing difficulty despite showing minimal stenosis on his control oesophagogram. Eight of the 23 patients with grade 2b burns had TPN. They were discharged after 3–27 days (mean

12 days). Six of them had developed strictures on fol-

low-up (26% of grade 2b patients). All the strictures

developed around post-burn week 3 except for one pa- tient who developed a stricture in post-burn week 5 despite a normal barium meal in post-burn week 3. Gastrostomy was performed on four patients both for feeding and dilatation purposes. Three of these had oesophageal perforations during dilatation that re- sponded to chest tube drainage. One of them is still on the dilatation program. Two patients with local necrosis grouped as having grade 3a burns had a smooth recovery, and neither developed strictures. They did not receive TPN during their hospitalisation. The only patient with grade 3b burns developed oesophageal stricture and had TPN for 7 days. She was lost to follow-up while she was on the dilatation pro- gram. The patient with severe grade 3 oral burns developed an oesophageal stricture during his hospitalisation. Be- cause of the severity of his oral burns, oesophagoscopy was not performed until stricture development. He had TPN for 15 days, and his stricture responded to two dilatations. Oesophageal strictures were seen after acid ingestion in two patients, alkali ingestion in four patients, and other caustics (unknown house cleaners) in two patients. Two patients with alkali ingestions needed long-term dilatations, and one of these is still on the dilatation program, post-burn month 8. One patient, who had acid ingestion with grade 1 oesophageal burn and erosive gastritis, developed pylo- ric obstruction. He had gastroduodenostomy on post- burn day 22. Median follow-up was 9 months (7 months–5 years). There was no mortality in the series.

Discussion

The degree and the extent of a corrosive lesion depend on several factors: the nature of the caustic substance, its concentration, the quantity swallowed, and the length of time the substance was in contact with the tissues. Alkalis dissolve the tissue and therefore penetrate more deeply, whereas acids cause a coagulative necrosis that limits their penetration. The lesions caused by alkali injury occur in three phases. First is the acute necrotic

Table 4 Results according to the grade of burns and type of ingestion

Oesophageal burn (no. of patients)

Acid

Alkali

Other

Oesophageal

Gastric

Gastric

ingestion

ingestion

caustic

stricture

burns

outlet obstruction

1 (24)

11

9

4

0

1

1

2a (21)

10

9

2

1

2

0

2b (23)

9

13

1

6

4

0

3a (2)

1

1

0

0

0

0

3b (1)

1

0

0

1

? a

0

Total (71)

31

32

7

8

7

1

phase, lasting 1–4 days after injury. During this period, coagulation of intracellular proteins results in cell necrosis, and the living tissue surrounding the area of necrosis develops an intense inflammatory reaction. Second is the ulceration and granulation phase, starting 3–5 days after injury. During this period the superficial necrotic tissue sloughs, leaving an ulcerated, acutely in- flamed base, and granulation tissue fills the defect left by sloughed mucosa. This phase lasts 10–12 days, and it is during this period that the oesophagus is the weakest. Third is the phase of cicatrisation and scarring, which begins the 3rd week following injury [16]. During this period the previously formed connective tissue begins to contract, resulting in narrowing of the oesophagus. The development of oesophageal strictures is related to the depth and circumferential nature of the corrosive injury [12]. These two factors were used in grading the oesophageal burns in the series, and treatment was planned accordingly [11]. The presence and degree of oral burns were not predictive for the presence and the degree of the oesophageal burns. Because all the inges- tions were accidental, a small number of patients had grade 3 burns. The main goal of the acute-phase treat- ment was to prevent oesophageal strictures. There is no standard acute management protocol for caustic burns. Several management protocols have been proposed, including the use of antibiotics, systemic ste- roids, antacids, and proton pump inhibitors. In some series, antibiotic therapy was used when definite infec- tion is present [6], whereas others use it for all burns [3, 13]. Long-term steroids have been tried and found effi- cient in some series [3], whereas others found them useless [1, 9 11, 12, 15]. Single-dose steroids were given to decrease oedema in some series [5]. Antacids and H2 blockers have been used both prophylactically or only in the presence of reflux [12]. Some authors delay oral feeding until swallowing occur [12], while others begin to feed the child following endoscopic control [10]. Many authors advocate nasogastric or nasoduodenal tubes for enteral nutrition [5, 11, 12] or insert nasogastric tubes as their standard management [4]. The incidence of stric- ture formation also differs between series, varying from 5% to 40% [3, 8, 14]. As with skin burns, factors influencing the inflam- matory reaction in the first few days and collagen syn- thesis later may play a role in future stricture development [6]. Infection is one of these factors. It is the experience of our department that because the oesophageal mucosa is fragile after the burn, food sticks to the mucosa even if the child can swallow well and is given oral foods. This is probably due to both the mucosal injury and the failure of peristalsis. These food impactions may lead to the colonisation of the oesoph- ageal wall and increase the inflammatory reaction that began with the burn, thus increasing collagen produc- tion. Colonisation may also increase the tissue damage and lead to deeper tissue destruction. Therefore, we delayed oral feeding until the time that the collagen production began to decrease, which was approximately

827

after the end of the 1st week, and we added antibiotics to the treatment. A single-dose steroid was given to de- crease the oedema due to the burn and thus help swal- lowing. Antibiotics were added both for the presence of infected oral burns and for oesophageal burns. There is no information in the literature about the colonisation of the oesophagus after a caustic burn, so antibiotics that cover the oral flora were chosen. To irrigate the oesophagus, drinking water was permitted as soon as the child began swallowing. No nasogastric tube was used for the purpose of feeding or stenting. H2 receptor blockers were given to patients for medical treatment of possible gastro-oesophageal reflux [13] and erosive gas- tritis due to acid ingestions. The stricture rate in our series was 17% for all pa- tients with grades 2 and 3 burns taken together (grade- s 2a, 2b, 3a, and 3b together) and 26% for patients in grade 2b taken alone. Alkali ingestions were associated with more oesophageal strictures, as described in the literature [5]. Only two of the eight patients who devel- oped oesophageal stricture had acid ingestion. The only patient with gastric outlet obstruction had acid inges- tion. Using this protocol, even if a stricture developed, we observed an early response to dilatations; currently, we have only one patient still on dilatation. No patient needed dilatations for more than 1 year (we have no information about the grade 3b patient who was lost to follow-up). The main goal in our acute-phase treatment was to keep the oesophagus as clean as possible in order to decrease the inflammatory reaction, which is compatible with the philosophy of wound healing. In one series, the authors proposed to insert nasogastric tubes for grades 2 and 3 burns and keep them for a month, and all their grades 2 and 3 patients developed oesophageal strictures [7]. Antibiotics, H2 receptor blockers, irriga- tion by water, and avoidance of any foreign bodies in the oesophagus were the main components of our treatment. Although we have no histopathological evidence show- ing decreased inflammation on the oesophageal wall, the clinical improvement was apparent. The parents de- scribed less pain and discomfort and felt happy to follow the protocol and give the medications to their children. Currently, another prospective study is planned with oral feeding beginning after weekly endoscopic controls of the oesophagus. We believe that further research is needed regarding the microbial flora and inflammatory reactions of the oesophageal wall to prevent the for- mation of strictures.

References

1. Anderson KD, Rouse TM, Randolph JG. (1990) A controlled trial of corticosteroids in children with corrosive injury of the esophagus. N Engl J Med 323:637–640

2. Ashcraft KW (2000) The esophagus. In: Ashcraft KW (ed) Pediatric surgery, 3rd edn. WB Saunders, Philadelphia, pp 325–

347

3. Bautista Casasnovas A, Estevez Martinez E, Varela Cives R, Villanueva Jeremias A, Tojo Sierra R, Cadranel S (1997) A

828

 

retrospective analysis of ingestion of caustic substances by chil- dren. Ten year statistics in Galicia. Eur J Pediatr 156:410–414

10. Marie JP, Dehesdin D, De Sevin E, Andrieu-Guitrancourt J (1992) Management of caustic burns of the esophagus in chil-

4.

Broto J, Asensio M, Soler Jorro C, Marhuenda C, Gil Vernet

dren. Ann Pediatr (Paris) 39:495–500

JM, Acosta D, Boix Ochoa J (1999) Conservative treatment of caustic esophageal injuries in children: 20 years of experience. Pediatr Surg Int 15:323–325

11. Millar AJW, Cywes S (1998) Caustic strictures of the esophagus. In: O’Neill JAO, Rowe MI, Grosfeld JL, Fonkalsrud EW, Coran AG (eds) Pediatric surgery, 5th edn. Mosby, St Louis, pp 969–

5.

Cecconello I, Zilberstein B, Pinotti HW (1995) Upper gas-

Wastell C, Nyhus LM, Donahue PE (eds) Surgery of the

679

trointestinal sequelae due to ingestion of corrosives. In:

esophagus, stomach and small intestine, 5th edn. Little,

12. Miller KA, Dudgeon DL (2003) Caustic esophageal injuries and perforations. In: Ziegler MM, Azizkhan RG, Weber TR (eds)

Operative pediatric surgery. McGraw-Hill, New York, pp 341–

Brown and Company, Boston, pp 318–326

347

¨

6.

Cohen IK, Diegelmann RF, Yager DR, Wornum III IL, Graham MF, Crossland MC (1999) Wound care and wound healing. In: Schwartz SI (ed) Principles of surgery, 7th edn. McGraw-Hill, New York, pp 263–296

13. Mutaf O, Genc¸ A, Herek O, Demircan M, O zcan C, Arı kan A (1996) Gastroesophageal reflux: a determinant in the outcome of caustic esophageal burns. J Pediatr Surg 31:1494–1495

14. Nunes AC, Romaozinho JM, Pontes JM, Rodrigues V,

7.

Huang YC, Ni YH, Lai HS, Chang MH (2004) Corrosive esophagitis in children. Pediatr Surg Int 20:207–210

Ferreira M, Gomes D, Freitas D (2002) Risk factors for stricture development after caustic ingestion. Hepatogast-

8.

Lamireau T, Rebouissoux L, Denis D, Lancelin F, Vergnes

roenterology 49:1563–1566

P, Fayon M (2001) Accidental caustic ingestion in children:

15. Ulman I, Mutaf O (1998) A critique of systemic steroids in the

is endoscopy always mandatory? J Pediatr Gastroenterol Nutr 33:81–84

management of caustic esophageal burns in children. Eur J Pediatr Surg 8:71–74

9.

Mamede RC, De Mello Filho FV (2002) Treatment of caustic ingestion: an analysis of 239 cases. Dis Esophagus

16. Warden GD, Heimbach DM (1999) Burns. In: Schwartz SI (ed) Principles of surgery, 7th edn. McGraw-Hill, New York, pp 223–

15:210–213

262

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.