Discussion

In fact, this patient exhibited the classic signs of Horner's syndrome: a loss of sympathetic innervation to the face from injury at various levels of the spinal cord or sympathetic chain. A basic understanding of the anatomy of the sympathetic division of the autonomic nervous system is required to understand the resulting signs. In the sympathetic division, cell bodies of preganglionic neurons are located in the lateral horn of the spinal cord (between the first thoracic level and the second or third lumbar level). These neurons send axons to paravertebral ganglia within the sympathetic chain. In the context of innervation to the head and neck, preganglionic fibres ascend to the superior cervical ganglion where they synapse onto postganglionic neurons. Axons from these neurons then innervate the head and neck. The clinical signs in this case consisted of partial ptosis (due to loss of sympathetic innervation to the levator palpebrae), meiosis (due to unopposed parasympathetic action on the sphincter papillae) and, as the lesion was below the superior sympathetic ganglion, anhydrosis. It remains unclear whether this case of Horner's was missed on first presentation or resulted from the blade remaining, and possibly moving, in the patient's neck over the 2-day period. Interestingly, this patient's transaxial stab was associated with a contralateral Horner's.

Horner's syndrome has multiple aetiologies, including malignant tumours (the most common cause) and benign tumours that are vascular, infectious, iatrogenic (including chest-drain insertion, central venous access, thyroidectomy and carotid angiography) and traumatic (including penetrating, blunt and birth related). Horner's has been previously described in the context of penetrating trauma, though isolated Horner's is uncommon. However, review of the literature suggests that this is the first reported case of a patient who presents with Horner's in the context of a penetrating neck injury with the foreign body still in situ.

The assessment and management of penetrating neck injuries is complex. The neck contains multiple structures in close proximity and surgical exploration can be difficult, requiring a thorough understanding of the relevant anatomy. There is no debate about the need to surgically explore a patient who is unstable or with hard signs of vascular injury. However, the management of patients with either no signs or ‘soft’ signs of clinical injury remains controversial. Some surgeons advocated mandatory exploration of all wounds penetrating the platysma, irrespective of signs and symptoms. They argue that physical signs are unreliable and that morbidity from negative exploration is negligible and preferable to missed injury and its complications. This approach is associated with less invasive diagnostic studies (and consequently reduced costs) and short hospital stays.1 Nonetheless, this is also associated with a significant rate of negative explorations (57% in the study described previously1 and higher in others), and recent evidence increasingly supports selective nonoperative management of penetrating neck injuries, relying on careful clinical examination and appropriate use of diagnostic imaging.2

It is important to remember that availability of diagnostic tools, costs and quality of follow-up are factors that are likely to influence a surgeon's approach to penetrating neck trauma. Thus, in some cases mandatory exploration may still be appropriate.

This case is clearly more complicated as the knife blade remained in situ. Two similar cases have been reported (neither with associated neurology).3 4 Both used emergency CT angiography to exclude direct involvement of major vessels. In the case of a straight blade, this was easily removed via the entry site.3 In the more complicated case of a barbed spear, exploration was commenced via a median sternotomy to allow vascular control if required. The spear was then removed via the entry site while retracting major structures under direct vision.4