2017-18 HSC Section 3 Green Book

m a d s e n e t a l c t a f o r a e r o d i g e t s i v e t r a c t i n j u r y f o l l o w i n g p e n e t r a t i n g n e c k i n j u r y

were localized to the oral cavity, 4 pharynx, 18 cervical esoph- agus, 8 and thoracic esophagus. 2 In two patients with pharyngeal injuries, the trajectory of the penetrating object was transpharyngeal on the CTA, this was secondary to a GSW in one patient and an SW in the other. In all patients with a DTI, except one with a surgically insignificant injury to the oral cavity from a stab through the floor of the mouth, did the CTA studies demonstrate deep surgical emphysema. Another 126 patients had CTA studies positive for this finding but without being diagnosed with a DTI. The remaining 219 studies did not demonstrate deep surgical emphysema. None of these 219 patients required readmission for a missed DTI. Deep surgical emphysema therefore had a sensitivity of 97.4% (37/[37 þ 1]) in identifying a DTI. This increases to 100% when only assessing for clinically relevant DTIs and excluding in- juries to the oral cavity. The specificity was as low as 63.5% (219/[219 þ 126]) and the PPV only 22.7% (37/[37 þ 126]). The NPV of deep surgical emphysema in excluding a DTI was 99.5% (219/[219 þ 1]) but also increased to 100% when excluding oral cavity injuries. Table 3 demonstrates these patients. Four of these 38 patients were proven to have concomitant airway injury.

Eighteen patients had odynophagia of which two also had an airway injury. Seven patients had palpable subcutaneous emphysema of which two had a concomitant airway injury. No patient, not immediately explored or imaged with CTA on admission, admitted for serial neck examinations developed late findings of DTI in our series.

The significance of surgical emphysema on CTA in patients with airway injury

A total of 11 (2.9%) patients had a confirmed airway injury. Six were confirmed on neck exploration, four by bronchoscopy and one by direct visualization of a blowing wound from a self-inflicted laceration to the neck. Four of these 11 patients also had a concomitant DTI. All had deep surgical emphysema on CTA, except from the patient with the blowing wound who was a late referral. He only had a CTA study done 10 d post- injury to exclude occult vascular injury. On arrival, the wound was approximated with sutures, and an occlusive dressing was applied as would be done with a tracheostomy wound. The patient experienced no complications relating to his airway. Another 153 patients without a confirmed airway injury had deep surgical emphysema demonstrated on CTA studies. The sensitivity, specificity, PPV, and NPV of deep surgical emphysema in identifying airway injuries were therefore 91.0% (10/[10 þ 1]), 63.5% (219/[219 þ 126]), 6.1% (10/ [10 þ 153]), and 99.5% (219/[219 þ 1]), respectively. A further eight patients, with PNI secondary to an SW, had anairway injury exclusivelydiagnosedon thebasisof theirCTA findings. In one of these patients, the trajectory was through the thoracic trachea, and there was deep surgical emphysema.

The significance of clinical signs and symptoms of a DTI

Of these 38 patients with a confirmed DTI 29 (76%) had clinical findings suggestive of an ADTI, seven had inadequately recorded clinical findings and two an impaired level of con- sciousness, which precluded clinical examination. The two most common clinical symptoms and findings were odyno- phagia and/or dysphagia and subcutaneous emphysema.

Table 3 e Significance of surgical emphysema on CTA in identifying DTI. CT findings Patients with DTI

Patient without DTI

Any DTI

Oral cavity

Pharynx

Cervical esophagus

Thoracic esophagus

CTA suggestive of injury total

37

3

21

11

2

126

Deep cervical emphysema

37

3

21 (two with missile tract identified as through the pharynx)

11

2

126

Mediastinal

20

1

8

9

2

68

emphysema

CTA not suggestive of injury

1

One (stab through the floor of the mouth)

0

0

0

219 (35 of these CTA could not be reviewed for emphysema)

Total

38

4

21

11

2

345

Sensitivity

97.4% (37/(37 þ 1])

75% (3/[3 þ 1])

100% (21/21)

100% (11/11)

100% (2/2)

Specificity

63.5% (219/ [219 þ 126])

Positive predictive value (PPV) Negative predictive value (NPV)

22.7% (37/[37 þ 126])

2.3% (3/[3 þ 126])

14.3% (21/[21 þ 126]) 8.0% (11/[11 þ 126]) 1.6% (2/[2 þ 126])

99.5% (219/ [219 þ 1])

CTA ¼ computed tomography angiography; DTI ¼ digestive tract injury.

50