Prolonged brachial plexus neuropathy: a rare complication following protracted endometriosis surgery in Lloyd-Davies position

Background

Brachial plexus injury is one of the most common peripheral nerve injuries related to surgical positioning with an estimated incidence of 0.16%.1 The prevalence of peripheral nerve injury associated with laparoscopy is unknown but estimates range from 0.03% to 1.19%.2 3 During laparoscopic surgery, peripheral nerve fibres may be compressed, stretched, entrapped or transected as a result of a combination of patient-specific risk factors such as body habitus and patient positioning.4

Commonly adopted surgical positions in laparoscopic gynaecology procedures include Trendelenburg and Lloyd-Davies positions. Friedreich Trendelenburg originally described his eponymous position as head-down tilting of an otherwise supine patient (classically with the patient’s legs on the assistant’s shoulders) but the Trendelenburg position nowadays refers to any head-down tilt, usually at 45°. Lloyd-Davies position is also known as Trendelenburg position with legs apart or ‘head-down lithotomy’. The latter is associated with fewer lower limb neuropathological side effects compared with lithotomy where the hips are almost fully flexed.5 Safe intraoperative patient positioning must balance optimal operating field access and visualisation with risk of neurological compromise and is the responsibility of the multidisciplinary surgical team.

Case presentation

A 38-year-old ex-smoker, with a body mass index (BMI) of 19, underwent elective laparoscopic surgery for severe endometriosis, with a total procedure time of 270 min. During this prolonged period, the patient was placed in Lloyd-Davies position with steep head-down tilt for most of the operation. The right arm was pronated, abducted at less than 90° and secured to an anaesthetic arm board. Bilateral shoulder supports were placed and the limb was strapped at the mid-upper arm level (figure 1). On awakening from anaesthesia, the patient reported inability to move her right arm associated with numbness.

Figure 1

Example of poor patient arm position during surgery.

Investigations

Initial neurological examination of the dominant right upper limb showed reduced tone, absent reflexes and 1/5 power on the Medical Research Council (MRC) scale. There was loss of all sensation below the level of the shoulder. Vascular supply was not impaired with findings of a warm hand, normal capillary refill time in all digits and easy palpation of the ulnar and radial pulses. There was minimal improvement of neurological function after 24 hours, and figure 2 shows that shoulder abduction, flexion and extension remained limited when assessed 3 days following surgery. MRI of the head and the brachial plexus was normal. Neurological examination at 1 week showed mild winging of the right scapula. Power in the muscles was as follows: deltoid (MRC grade 4/5), trapezius (4/5), latissimus dorsi (4−/5), biceps (4/5), triceps (4+/5); supraspinatus and infraspinatus and rhomboids (4/5); extensor digitorum (4/5), abductor digiti mini (4−/5), first dorsal interosseous and abductor pollicis brevis (3/5). There was patchy pinprick sensation throughout the right arm and reflexes remained absent.

Figure 2

Right-sided brachial plexus neuropathy. At 3 days post-surgery, the patient has limited shoulder flexion and abduction in her affected right arm.

Differential diagnosis

At the time of injury, the main clinical suspicion was traumatic brachial plexus nerve injury with the potential causes of injury secondary to patient positioning include:

• Steep head-down tilt of Trendelenburg.

• Arm hyperabduction.

• Arm fixed to an arm board, possibly causing tourniquet paralysis syndrome or nerve entrapment.

• The positioning of shoulder supports.

• Protracted operating time.

Differential diagnoses for brachial plexus disorders include:

• Infective.

• Infiltrative (brachial plexus neoplasm).

• Inflammatory.

• Paraneoplastic.

• Anatomical (a cervical rib).

Treatment

Specialist neurological advice was sought. Unfortunately, the peripheral neurophysiology study was significantly delayed due to the COVID-19 pandemic. Nevertheless, she was managed conservatively with physiotherapy-directed exercises and advised on limb support while sitting and mobilising. Initially, a broad arm sling was used to prevent further injury to the ipsilateral side. Neuropathic pain was managed with amitriptyline 10 mg at night. An occupational health assessment was sought to guide her phased safe return to work as a paediatric nurse.

Outcome and follow-up

At 6 weeks, there had been significant motor improvement with only minimal residual hypoaesthesia in the dermatomes of C5–C7. At 10 weeks, the electromyography (EMG) and nerve conduction study demonstrated a markedly attenuated right lateral antebrachial cutaneous response compared with the left. Upper limb sensory responses were normal, with no asymmetry between the two sides. Upper limb median and ulnar motor velocities were normal with no asymmetry in the distal compound muscle action potentials between the two sides. Needle EMG of right biceps revealed some mild neurogenic changes but no active denervation. The rest of the sampled muscles including trapezius, serratus anterior, rhomboid major, infraspinatus, deltoid, triceps, extensor digitorum, pronator teres, abductor pollicis brevis, abductor digiti mini and first dorsal interosseous were normal. The assessing neurophysiologist’s conclusion was of a ‘pan-brachial plexopathy which had improved both clinically and electrophysiologically, with a residual musculocutaneous neuropathy’.