Secondary Logo

Journal Logo

RFS – Clinical Vignette

Perioperative Weakness

Clinical Vignette and Review of Mechanism, Incidence, and Risk Factors

Ruiz-Peters, Michael MD; Winston, Paul MD, FRCPC; Krauss, Emily M. MD, MSc, FRCSC

Author Information
American Journal of Physical Medicine & Rehabilitation: October 2020 - Volume 99 - Issue 10 - p 970-973
doi: 10.1097/PHM.0000000000001402


A healthy 24-yr-old man was in a collision with a stationary vehicle while riding a toboggan on a ski hill, resulting in bilateral femur fractures, left tibia and fibula fractures, and a right pilon fracture. Medical, family, and psychosocial history was noncontributory. He underwent intramedullary nailing of bilateral femur fractures and the tibia fracture during a single operation. A blood pressure cuff for intraoperative monitoring was used on the left arm with bruising noted in the left cubital fossa postoperatively. All fractures were fixed in the same operation, with a total operative time of 7 hrs. Postoperatively, he presented with bilateral, left greater than right, upper limb weakness. Specifically, flaccid weakness in flexor pollicis longus (FPL), abductor pollicis brevis (APB) with weakness of flexor digitorum superficialis and flexor digitorum profundus (FDP) to digits I and II (Medical Research Council [MRC] grade 1 bilaterally), and MRC grade 1–2 pronation bilaterally. The muscles innervated by the radial and ulnar nerves, including FDP to digits III and IV, were of normal strength (MRC 5). He had no spasticity or upper motor signs. Left elbow flexion power and range of motion was reduced secondary to pain, and there was nonspecific, patchy hand numbness.


A reasonable initial differential diagnosis could have included the stroke, central cord injury, peripheral nerve injury, and unrecognized muscle or tendon injury on primary survey. A working diagnosis of intraoperative fat embolic stroke was preferred by the admitting service, given the left greater than right involvement and recent surgery. However, the lack of upper motor signs and focal presentation would have refuted this. Central cord injury, with upper great than lower weakness, could also be entertained especially given the traumatic nature of the event. More focal injuries, such as traumatic peripheral nerve or muscle/tendon injuries were low on the primary differential.


To better understand the possible cause of weakness in both hands, a magnetic resonance imaging scan of the brain was ordered. It showed findings consistent with multiple cerebral fat emboli, and the patient was believed to have an intraoperative stroke. However, physiatry consultation noted weakness in a lower, rather than upper, motor neuron pattern. Electromyography was then arranged 3 wks after the onset of weakness to allow a potential nerve injury to declare itself. The study revealed both an absent left median motor compound muscle action potential and median sensory potential, as well as a prolonged right median motor latency with low amplitude distally and no compound muscle action potential proximally. Electromyography showed bilateral median and anterior interosseous nerve (AIN) involvement. The left APB showed 1+ positive sharp waves and reduced recruitment. The left FPL had one unit firing and was otherwise silent. The left pronator teres (PT) revealed positive sharp waves, fibrillations, and reduced recruitment. The left flexor digitorum superficialis showed very reduced recruitment. The left FDP to digits I and II showed 2+ positive sharp waves, fibrillations, and reduced recruitment. The right APB showed fibrillation and reduced recruitment. The right FPL showed reduced recruitment. The right FDP showed reduced recruitment. The right PT was normal as were the radial and ulnar-innervated muscles. Both right and left arm magnetic resonance imaging showed edema in the bilateral brachialis muscles, around the median nerve, and along fascial tissue at the level of the bifurcation of the PT branch of the median nerve and the median nerves proper as they entered the forearm (Fig. 1).

Magnetic resonance imaging edema along the superficial fascia of the flexor-pronator mass (right arrow) and along the lacertus fibrosis (left arrow), which obscures visualization of the median nerve as it enters the forearm. The edema seen is likely from inflammation because of positioning, polytrauma, and repeated blood pressure monitoring.


The leading diagnosis was perioperative compression neuropathy. Potential management approaches included close monitoring and physical therapy, nerve decompression/release, nerve transfer, or tendon transfer. In this case, intervention proceeded from least to most invasive.

Ten weeks after injury and conservative management, his left side continued to have significant impairment while the right side was recovering clinically. Therapy consisted of occupational and physiotherapy focusing on improving independent movements such as transfers and self-care. He was nonweight bearing on his legs because of his lower limb fractures. There was no dedicated hand therapist in this setting; however, physical therapy was initiated in hospital and through a patient-performed home program to improve passive range of motion in the fingers, which were noted to be stiff. Active range of motion was encouraged, but strengthening activities were avoided during the initial phase.

When clinically assessed 8 wks after presentation, the patient was noted to have persisting severe motor and sensory deficits on the left side despite conservative management and therapy and evidence of recovery on the right side. The decision was made to pursue surgical decompression, suspecting ongoing compression neuropathy. The patient underwent a left median nerve decompression in the forearm. Intraoperatively, he was found to have a thick bicipital aponeurosis (lacertus fibrosus) and large perforating vessels across the median nerve, which were divided to decompress the nerve. Intraoperative electrical stimulation was performed on the motor branches to PT and to the AIN. The median nerve motor branches to PT had strong muscle activation with stimulation; however, when the AIN was stimulated there was minimal activation of FPL and other AIN-innervated muscles (index FDP). To optimize eventual distal motor recovery, a radial motor nerve to AIN end-to-side nerve transfer was performed (specifically using the radial motor nerve fascicles of extensor carpi radialis brevis as the donor nerve).

By 2 wks postoperatively, the AIN was showing early signs of clinical recovery. There was improved power to FPL (MRC 2) with slight weakness of FDP II and III (MRC 3−). A follow-up electromyography that was performed 4 mos after the first, three mos postoperatively, noted the right median motor now showed normal latency with slowed velocity, whereas the sensory showed a small response. The decompressed left median distal motor and sensory potentials at the carpal tunnel showed no response. The left PT electromyography was normal, whereas FDP I and II showed slightly reduced recruitment. The left APB showed 3+ denervation and reduced recruitment, while the right had 1+ denervation and slightly reduced recruitment. By 7 months, the patient had regained full function in both hands, including FPL, FDP II and III and APB function in the left hand (both MRC 4+) and improving sensation. Suspected multifactorial causes of his presentation include an anatomical variant with bilaterally pronounced lacertus fibrosis, prolonged surgery, major trauma with possible hypovolemia, and frequent intraoperative blood pressure monitoring causing perioperative compression neuropathy. Postoperatively, he noted that he had never been able fully extend his elbows, suggesting preexisting elbow tightness (possibly from the large bilateral lacertus fibrosis).


Perioperative nerve palsy is a well-documented complication, with pathophysiology related to biomechanical, biochemical, and microvascular injury.1

Biomechanically, tissue damage is from both elastic and plastic deformation via shear stress.2 There is a nonlinear relationship between shear forces and tourniquet pressure, meaning that a higher tourniquet pressure produces greater than expected shear damage.3 This effect was attenuated with the use of a larger cuff, which more equally distributed the forces. Biochemically, the decrease in circulation creates localized lactic acidosis and metabolic by-products, which leads to membrane disruption, and leukocyte activation. When macrovascular flow is returned there can be a “no-reflow phenomenon” due to this microvascular damage. Most importantly, this effect is compounded in microvascular diseases such as diabetes or hypertension.2

Anatomically, the distance of the median and radial nerves to points of osseous impingement decreases with elbow flexion, illustrating the potentially deleterious effect of prolonged flexion—much like the situation in our presented case. The incidence of perioperative neuropathy is estimated at 0.037%–0.26% of surgical cases, but this varied significantly depending on the procedure and patient population (Table 1).

Study type, study size, population composition, incidence, and recovery time

Tourniquet cuff duration and pressure are well-recognized risk factors, with a suggested 2-hr duration based on histologic and clinical data.2 If a longer duration is needed, 10 mins of reperfusion every hour can increase total tourniquet duration, but tissue injury was exacerbated if reperfusion was delayed until after 2 hrs. Regional limb cooling and exsanguination before cuff inflation may also decrease ischemic damage and edema.2 For tourniquet pressure, some studies were conducted with set pressures (300–500 mm Hg), whereas others recommended adjusting cuff pressures to 75 mm Hg above systolic,2 although there is no clear consensus in the literature.

Nonconsensus risk factors include sex, padding, and positioning. Two studies found no relationship to positioning, although these did illustrate that men had a higher rate of neuropathy, with 86% of neuropathy patients being male and a relative risk of 4.1.4,5 However, many studies suggest a positional factor,1,3,5,6 with one study illustrating that half of their study population had previous positional nerve palsies and asserted a correlation between positioning and the neuropathy.6 Though prevalent, patient padding during surgery has surprisingly little literature evaluating this intervention. A number of studies addressed this paucity and asserted the difficulty in evaluating padding in a standardized manner.1,2,4,6

Consensus risk factors include body mass index, diabetes, hypertension, preexisting subclinical neuropathy, and upper limb trauma. In both cadaveric and patient studies, extremes of weight exhibit nonideal pressure transmission throughout the tissue.3 Clinically, this correlated with a relative risk of 15 for obese patients and a relative risk of 3.7 for low body mass index patients.4 Microvascular diseases such as diabetes, smoking, and hypertension were theorized to decrease perfusion and impair reperfusion, making these patients more susceptible to injury.2 Diabetes was found to have a relative risk of 4.3,4 and a separate study found the hazard ratio for diabetes, hypertension, and tobacco use to be 2.4, 2.2, and 2.1, respectively, with 34% of neuropathy patients having hypertension. Multiple studies showed evidence of preexisting neuropathy in asymptomatic patients,4,6,7 as well as cases of patients who, despite proper padding and positioning, developed contralateral neuropathy during procedures attempting to fix the initial postsurgical palsy.4 Similar to our clinical vignette, there is also higher incidence of neuropathy in the upper limbs, even when the procedure itself is in an anatomically remote location.1,4,6,8

Clinical recovery after perioperative neuropathy is inconsistent in the literature. For ulnar neuropathies, two studies reported a bimodal distribution in recovery times where patients would either recover within 6 wks or have persistent symptoms at 2 or 5 yrs.5,6 A similar study reported 53% recovery within 1 yr, with the other half having persistent symptoms.4 As radial and median neuropathies are less common in the perioperative setting, one study provided a group outcome measure of all neuropathies recovering in 1 yr for 81/82 patients.8 From the surgical literature, early surgical decompression of compression neuropathies is recommended once the severity of the injury is established (generally approximately 12 wks after injury), to maximize recovery. This practice is supported by animal studies reporting restoration of neurovascular blood flow with early surgical decompression.9 In addition, experts recommend early nerve transfer surgery in patients with absent or poor recovery after nerve injury at the time of decompression as an alternative to traditional nerve reconstruction techniques such as tendon transfers.10


Despite the numerous causes for perioperative weakness, compression neuropathy should remain on the differential diagnosis especially when the following risk factors are identified: upper limb weakness despite remote surgery, tourniquet duration, microvascular diseases, upper limb injuries, males, and body mass index extremes. The overall incidence ranges widely, and symptoms can resolve in days or become chronic. Nerve decompression should be considered at 12 wks after injury if insufficient clinical improvement occurred. The addition of a nerve transfer can be considered to augment recovery. Preoperatively, physicians should recognize patients who have a number of risk factors and give particular attention to limb positioning, patient repositioning, padding, reperfusion timing and exsanguination, as well as appropriate tourniquet cuff size, placement, pressure, and duration. Testable features of this case include an approach to weakness in operative/trauma patients, as well as risk factors, management, treatment, and prognosis of compressive neuropathies.


I would like to thank Dr. Paul Winston and Dr. Emily Krauss for their mentorship, help in selecting this topic, and feedback on its production. I would also like to thank Valerie Dupuis for her assistance in gathering articles, and Dr. Rebecca Warburton and Dr. Daniel Burd for their assistance in reviewing this article.


1. Welch MB, Brummett CM, Welch TD, et al.: Perioperative peripheral nerve injuries: a retrospective study of 380,680 cases during a 10-year period at a single institution. Anesthesiology 2009;111:490–7
2. Pedowitz RA: Tourniquet-induced neuromuscular injury. Tourniquet-induced neuromuscular injury. A recent review of rabbit and clinical experiments. Acta Orthop Scand Suppl 1991;245:1–33
3. Graham B, Breault MJ, McEwen JA, et al.: Perineural pressures under the pneumatic tourniquet in the upper extremity. J Hand Surg Br 1992;17:262–6
4. Warner MA, Warner ME, Martin JT: Ulnar neuropathy: incidence, outcome, and risk factors in sedated or anesthetized patients. Anesthesiology 1994;81:1332–40
5. Warner AM, Warner OD, Matsumoto YJ, et al.: Ulnar neuropathy in surgical patients. Anesthesiology 1999;90:54–9
6. Lee CT, Espley AJ: Perioperative ulnar neuropathy in orthopaedics: association with tilting the patient. Clin Orthop Relat Res 2002;396:106–11
7. Alvine FG, Schurrer ME: Post-operative ulnar nerve palsy. J Bone Joint Surg 1987;69A:255–9
8. Middleton KWD, Varian JP: Tourniquet paralysis. ANZ J Surg 1974;44:124–8
9. Jung J, Hahn P, Choi B, et al.: Early surgical decompression restores neurovascular blood flow and ischemic parameters in an in vivo animal model of nerve compression injury. J Bone Joint Surg Am 2014;96:897–906
10. Mackinnon SE: Future perspectives in the management of nerve injuries. J Reconstr Microsurg 2018;34:672–4

Paresis; Peripheral Nerve Injury; Stroke; Perioperative Period

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.