• 
• 

Three Part Question

In [patients who are suspected to have fish or chicken bones impacted in the oesophagus] is [computerised tomography more effective] than [plain radiography at diagnosis]?

Clinical Scenario

A 60 year old man attends the ED complaining that a fish bone has got stuck in his throat. Clinical examination rules out impaction within the pharynx so you are concerned that the bone has become impacted within the oesophagus. Prior experience tells you that oesophageal abrasions secondary to ingested bones can often mimic impaction, that rigid oesophgoscopy (the definitive investigation) carries a significant mortality and morbidity rate, and that the most readily available non-invasive investigations, lateral neck and chest x-rays, are often unreliable. You wonder whether a CT scan of the neck would be a more accurate non-invasive tool?

Search Strategy

Medline 1966 – week 3, February 2006. Limit to human and English.
({exp. Tomography, Spiral Computed/or exp. Tomography, X-ray computed/or CT. mp.} AND [{fish bone. mp.} OR {foreign body. mp. OR exp. Foreign bodies}] AND {exp. Esophagus/or oesophageal. mp.})

Search Outcome

66 papers found of which 62 were irrelevant or of insufficient quality. The remaining four papers have been systematically reviewed.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Akazawa Y et al 2004 Japan	76 patients with suspected fish bone ingestion. All patients underwent lateral neck x-ray and CXR, followed by CT.	Prospective, non-randomised study.	Sensitivity	CT: 100%, XR: 64.5%	XR sensitivity miscalculated by authors as 54.8%.
			Specificity	CT: 100%, XR: 95.6%
			Positive predicitive value	CT: 100%, XR: 90.9%
Eliashar R et al 1999 Israel	45 patients with suspected fish and chicken bone ingestion. All patients underwent lateral neck x-ray and CXR, followed by CT.	Prospective, non-randomised study.	Sensitivity	CT: 100%, XR: 55.2%
			PRIMARY OUTCOME: Resting LVEF at 6 wks	51% EF in immediate group v/s 50.1%EF in deferred;p 0.22 no significant difference
			Specificity	CT: 93.7%, XR: 100%
			PRIMARY OUTCOME: Exercise LVEF on hospital discharge	50.7% EF in immediate group v/s 50.3%EF in deferred;p 0.60 no significant difference
			Positive Predicitive Value	CT: 96.7%, XR: 100%
			PRIMARY OUTCOME: Exercise LVEF LVEF at 6 wks	54.3% EF in immediate group v/s 54.1%EF in deferred;p 0.86 no significant difference
			SECONDARY OUTCOME:Death within 6 days,at 6wks and 1 yr	2.4%, 3.6% and 4.8% in immediate group v/s 2.4%, 3.5% and 5% in deferred;p 0.98, P 0.91 and p=0.87 - no significant difference
Watanabe K et al 1998 Japan	32 patients with suspected fish bone ingestion. All patients underwent lateral neck x-ray and CXR, but only 11 underwent CT.	Prospective, non-randomised study.	Sensitivity	CT: 100%, XR: 44%	Selection bias – why only 11 patients for CT? Low study numbers for CT. Statistics were not reported by authors (they have been calculated by ourselves).
			Early detection of bacteraemia	No numerical data in results. "Elevated levels of CRP in serum did not differ significantly among patients with bacteremia, patients infected but without bacteremia, and patients who were apparently ill enough for infection to be considered and blood cultures taken, but in whom no infection could be found. Furthermore, some Furthermore, some patients with bacteremia had levels of CRP in the normal range and therefore therefore little reliance can be placed on a normal level of CRP as an indication that bacteremia is absent in an ill patient."
			Specificity	CT: N/A, XR: 100%
			Positive Predicitive Value	CT: 100%, XR: 100%
Braverman I et al 1993 Israel	13 patients with suspected fish and chicken bone ingestion. All patients underwent lateral neck x-ray and CXR, followed by CT.	Prospective, non-randomised study.	Severity of infection	At < or =24h, the AUC(ROC)s of PLA2-II and PCT were superior to those of CRP.	Low study numbers. One patient underwent CT after oesophagoscopy, & one did not have an XR – both have been excluded form the results. Statistics were not reported (calculated by ourselves again).
			Sensitivity	CT: 100%, XR: 80%
			Specificity	CT: 100%, XR: 50%
			Positive Predictive Value	CT: 100%, XR: 88.9%

Comment(s)

The studies reviewed above clearly show that CT of the neck is an extremely accurate, non-invasive diagnostic tool with a high PPV. However, of the 58 patients in the four series with positive X-Ray findings, there were only 3 false-positives. Thus, disregarding study 4 (which is appears to be a small-scale pilot study for Study 2 with skewed results), it is unsurprising that Studies 1-3 all still recommend plain X-ray as the initial radiographic screening tool. Positive results, which include soft tissue changes, warrant oesophagoscopy, while negative results should lead to a CT scan of the neck. Bearing in mind that there was 100% sensitivity amongst the 144 patients undergoing CT, and that there was just one false positive amongst the 80 patients with positive results, only visualisation of an actual fish or chicken bone should result in an oesophagoscopy at this point. Similarly, a negative CT scan should confidently exclude fish and chicken bones.

Clinical Bottom Line

CT is more effective than plain radiography at identifying and excluding impacted oesophageal fish and chicken bones. However, plain radiography is also specific enough for positive results to warrant oesophagoscopy without any further imaging, and should thus continue being utilised as the first line radiological investigation.

References

1. Akazawa Y, Watanabe S, Nobukiyo S, Iwatake H, Seki Y, et al. The management of possible fishbone ingestion. Auris, Nasus, Larynx 2004; 31: 413-16.
2. Eliashar R, Dano I, Dangoor E, Braverman I, Sichel J-Y. Computed tomography diagnosis of esophageal bone impaction: a prospective study. The Annals of Otology, Rhinology, & Laryngology 1999; 108: 708-10.
3. Watanabe K, Kikuchi T, Katori Y, Fujiwara H, Sugita R, et al. The usefulness of computed tomography in the diagnosis of impacted fish bones in the oesophagus. The Journal of Laryngology and Otology 1998; 112: 360-64.
4. Braverman I, Gomori JM, Polv O, Saah D. The role of CT imaging in the evaluation of cervical esophageal foreign bodies. The Journal of Otolaryngology 1993; 22 (4): 311-14.