Ultrasound-guided Percutaneous Cryoneurolysis for the Treatment of Pain after Traumatic Rib Fracture: A Randomized, Active-controlled, Participant- and Observer-masked Study

Abstract

Background: Traumatic rib fractures are associated with pain lasting weeks to months and a decreased ability to inspire deeply or cough to clear secretions. Ultrasound-guided percutaneous cryoneurolysis involves reversibly ablating peripheral nerve(s) using exceptionally low temperature with a transdermal probe, resulting in a prolonged nerve block with a duration measured in months. The authors hypothesized that cryoneurolysis would improve analgesia and inspired volume after rib fracture.

Methods: Adults with one to six traumatic rib fractures were randomized to either active cryoneurolysis and sham peripheral nerve block or sham cryoneurolysis and active peripheral nerve block in a participant/observer-masked fashion. The primary endpoint was the maximum inspired volume the day after the procedure as measured with an incentive spirometer.

Results: The day after the procedure, the unadjusted median [interquartile range] maximum inspired volume for participants who received cryoneurolysis (n = 11) was 2,250 ml [1,500, 2,500 ml] versus 1,300 ml [750, 2,500 ml] for peripheral nerve block (n = 9, mean difference, 496; 95% CI, -428 to 1,420; t test P = 0.269). When adjusted for covariates ( e.g. , baseline lung volume), the cryoneurolysis group had an estimated 793 ml greater mean volume than peripheral nerve block (95% CI, 273 to 1,312 ml; analysis of covariance P = 0.005). Improvement from baseline in maximum inspired volume for cryoneurolysis was 1,000 ml [1,000, 1,375 ml] versus 300 ml [0, 1,000 ml] for peripheral nerve block ( t test P = 0.002). This was equivalent to an improvement over baseline of 100% [90%, 188%] for cryoneurolysis versus 30% [0%, 50%] for peripheral nerve block ( t test P = 0.003). Average daily pain scores were generally lower for the cryoneurolysis group throughout the first month. Total cumulative oxycodone equivalents were 5 mg [0, 13 mg] for cryoneurolysis versus 45 mg [43, 135 mg] for peripheral nerve block ( t test P = 0.013).

Conclusions: Ultrasound-guided percutaneous cryoneurolysis improves maximum inspired lung volume while concurrently decreasing pain and opioid consumption after traumatic rib fracture. These results should be considered preliminary, requiring confirmation with a trial including a larger sample size.

Fig. 1.

Ultrasound images of percutaneous cryoneurolysis of an intercostal nerve (used with permission from Wolters Kluwer).

Fig. 2.

Consolidated Standards of Reporting Trials diagram.

Fig. 3.

Effects of percutaneous cryoneurolysis on the increase in maximum inspiratory volume and pain after traumatic rib fracture. The maximum inspired volume was measured using a handheld incentive spirometer based on the American Association of Respiratory Care clinical practice guidelines, and the best of three measurements was recorded as the maximum inspired volume. For the panels involving improvement from baseline, data presented are the specific time point relative to baseline measured immediately before the intervention on day 0. Day 0 in the graphs was measured immediately after the intervention. Pain severity during incentive spirometer use was measured using a numeric rating scale with 0 equivalent to no pain and 10 being the worst imaginable pain. Data expressed as median (dark horizontal bars) with 25th to 75th percentiles (box), 10th to 90th percentiles (whiskers), mean (diamonds), and outliers below the 10th or above the 90th percentiles (circles). P values are derived from a mixed model for repeated measures with covariates for the baseline value (where collected for that outcome), time, and the group-by-time interaction.

Fig. 4.

Effects of percutaneous cryoneurolysis on pain and pain’s interference in functioning after traumatic rib fracture. Pain severity was measured using a numeric rating scale with 0 equivalent to no pain and 10 being the worst imaginable pain for the 24 h before data collection. Regarding the Brief Pain Inventory, pain interference indicated using a numeric rating scale of 0 to 70, with 0 and 70 equivalent to no and maximal interference, respectively. Data expressed as median (dark horizontal bars) with 25th to 75th percentiles (box), 10th to 90th percentiles (whiskers), mean (diamonds), and outliers below the 10th or above the 90th percentiles (circles). P values are derived from a mixed model for repeated measures with covariates for the baseline value (where collected for that outcome), time, and the group-by-time interaction.

Fig. 5.

Effects of percutaneous cryoneurolysis on the maximum pain level experienced in the entire year after traumatic rib fracture. Pain severity was measured using a numeric rating scale with 0 equivalent to no pain and 10 the worst imaginable pain. Post hoc analysis results (left) expressed as median (dark horizontal bars) with 25th to 75th percentiles (box), 10th to 90th percentiles (whiskers), mean (diamonds), and outliers below the 10th or above the 90th percentiles (circles); and (right) categorized as mild (numeric rating scale less than 4), moderate (numeric rating scale 4 to 7), and severe (numeric rating scale greater than 7) pain. On the right, the cryoneurolysis percentages do not total 100% due to a rounding error. P values are derived from a mixed model for repeated measures with covariates for the baseline value (where collected for that outcome), time, and the group-by-time interaction. PNB, peripheral nerve block.

Fig. 6.

Effects of percutaneous cryoneurolysis on opioid consumption after traumatic rib fracture. Oxycodone is a synthetic opioid and presented in milligrams. During this period, cryoneurolysis lowered cumulative opioid use by 89%, with the treated group using 5 mg [0, 13 mg] of oxycodone compared with 45 mg [43, 135 mg] in controls (P = 0.013). Regarding the risk of requiring any opioids at each time point, post hoc analysis results are presented as the percentage of each treatment group consuming any opioids at each time point. Data expressed as median (dark horizontal bars) with 25th to 75th percentiles (box), 10th to 90th percentiles (whiskers), mean (diamonds), and outliers below the 10th or above the 90th percentiles (circles). P values are derived from a mixed model for repeated measures with covariates for the baseline value (where collected for that outcome), time, and the group-by-time interaction.