2017-18 HSC Section 3 Green Book

▶ Table 2 Details of endoscopic grading on days 1 and 5 and subsequent development of esophageal and antropyloric stenosis.

Esophageal stricture present (n = 13)

No esophageal stricture (n = 49)

Antropyloric stenosis present (n = 20)

No antropyloric stenosis (n = 42)

Day 1

32 (97.0 %)

1 (3.7 %)

26 (96.3 %)

▪ Mild

1 (3.0 %)

▪ Severe

12 (41.4 %)

17 (58.6 %)

19 (54.3 %)

16 (45.7 %)

Day 5

41 (97.6 %)

4 (9.1 %)

40 (90.9 %)

1 (2.4 %)

▪ Mild

16 (88.9 %)

2 (11.1 %)

▪ Severe

12 (60.0 %)

8 (40.0 %)

ROC curves were plotted using severe injury on day 1 EGD and day 5 EGD as predictors of subsequent esophageal ( ▶ Fig. 4 ) or antropyloric stenosis ( ▶ Fig. 5 ). Severe injury on day 5 gave higher AUCs than severe injury on day 1 for the prediction of subsequent stenosis, in both the esophageal and antropyloric areas (0.88 vs. 0.79; P < 0.001). Discussion The present study highlights the importance of relook endos- copy to grade the mucosal injury in patients with caustic inges- tion. Upper GI evaluation was successfully done without any complications in 62 patients on both day 1 and day 5 after caus- tic ingestion. With relook endoscopy on day 5, nine patients (14.5 %) who had been graded as ≥ 2b esophageal injury on day 1 were downgraded. Similarly, 17 of the patients with se- vere gastric injury on day 1 (27.4 %) were downgraded on day 5 EGD. Evidence of severe injury on day 5 EGD was a better pre- dictor of the subsequent development of symptomatic esopha- geal and gastric cicatrization than the findings on day 1 EGD. While early EGD ( < 24 hours) has been the accepted norm for the evaluation of mucosal injury to the upper GI tract, some re- ports have questioned the reliability of EGD findings [27 – 30]. Chirica et al. [28] reported that 15 % of esophagectomies and 13.3 % of explorative laparotomies were done unnecessarily based upon endoscopic findings. They found that early endos- copy findings overestimated the severity of injuries. This over- estimation of the severity of the mucosal injury at EGD can be attributed to mucosal and submucosal hemorrhages, which can be misinterpreted as charring (grade 3 injuries). Extensive mu- cosal and submucosal edema can also create confusion in esti- mating the circumferential extent of the mucosal injury. More- over, during early endoscopy (within hours after caustic inges- tion), detailed evaluation is often limited by the presence of food residue and residual caustic material in the stomach, smearing of the mucosa with blood, and retching or peristalsis. We hypothesized that a relook endoscopy on day 5 would overcome all of these factors and would also allow better eval- uation after any mucosal/submucosal edema and hemorrhages had subsided. Our data support this contention, as there was a downgrading of both esophageal and gastric injury on the day 5 EGD.

The optimum timing of endoscopy has been a matter of de- bate. Most authors recommend early endoscopy (within 12 – 24 hours of caustic ingestion) [16]. Some authors suggest however that endoscopy may be done anytime within the first 96 hours, but that it should be avoided from 5 to 15 days after caustic in- gestion [8, 31]. It was believed that during the first 3 weeks after caustic ingestion, the tensile strength of the healing tis- sue was low as collagen deposition might not begin until the second week [10]. However, with better and slimmer endo- scopes, many endoscopists have demonstrated the safety of endoscopy at any time between the first week and 4 weeks after caustic ingestion [1, 15, 20]. Mauli et al. [32] have suggested that, in patients with severe oropharyngeal burns, endoscopy may be deferred up to 7 days to allow acute edema to subside. Cabral et al. [1] repeated endoscopy 1 week after caustic ingestion in 315 patients and based their management algorithm on endoscopic findings. Boskovic and Stankovic [15] reported their experience with 176 children in whom they repeated the endoscopy 5 – 15 days after the first one. Tohda et al. [20] repeated endoscopy 2 – 5 days after the first one in patients with grade 3 injuries to eval- uate progression/healing. All these studies have confirmed the safety of currently used endoscopes in patients with acute caustic ingestion. However, none of the authors reported whether the endoscopic findings on day 1 were any different from those on subsequent endoscopies. Although we did not use CO 2 during EGD in our patients, it must be emphasized that the use of CO 2 will enhance the safety of the procedure [33 – 35]. We chose day 5 for the repeat EGD based on our previous experience [23]. It would be interesting to note the changes in EGD findings on days 2, 3, 4, and 5 to determine what the ideal time to repeat EGD is; however, such a study would raise ethical and logistical issues. There is a linear correlation between the grading of mucosal injuries on EGD and the development of strictures/cicatrization, as was first demonstrated by our research group [8]. Of a total of 14 patients with 2b injury, 10 (71.4 %) developed strictures, while 26 of the 31 patients with grade 3 injuries (83.9 %) devel- oped strictures. Subsequent studies have also confirmed these findings [15, 20]. Boskovic and Stankovic [15] reported that two-thirds of their 27 patients with severe injury developed strictures. Tohda et al. [20] reported stricture formation in 11 of their 21 patients (52.4 %) with ≥ 2b injury. However, Cheng

Kochhar Rakesh et al. Relook endoscopy predicts … Endoscopy

216