Tourniquet Use in Wide-Awake Carpal Tunnel Release

Abstract

Background: Carpal tunnel syndrome is a common cause of upper extremity discomfort. Surgical release of the median nerve can be performed under general or local anesthetic, with or without a tourniquet. Wide-awake carpal tunnel release (CTR) (local anesthesia, no sedation) is gaining popularity. Tourniquet discomfort is a reported downside. This study reviews outcomes in wide-awake CTR and compares tourniquet versus no tourniquet use. Methods: Wide-awake, open CTRs performed from February 2013 to April 2016 were retrospectively reviewed. Patients were divided into 2 cohorts: with and without tourniquet. Demographics, comorbidities, tobacco use, operative time, estimated blood loss, complications and outcomes were compared. Results: A total of 304 CTRs were performed on 246 patients. The majority of patients were male (88.5%), and the mean age was 59.9 years. One hundred patients (32.9%) were diabetic, and 92 patients (30.2%) were taking antithrombotics. Seventy-five patients (24.7%) were smokers. A forearm tourniquet was used for 90 CTRs (29.6%). Mean operative time was 24.97 minutes with a tourniquet and 21.69 minutes without. Estimated blood loss was 3.16 mL with a tourniquet and 4.25 mL without. All other analyzed outcomes were not statistically significant. Conclusion: Operative time was statistically longer and estimated blood loss was statistically less with tourniquet use, but these findings are not clinically significant. This suggests that local anesthetic with epinephrine is a safe and effective alternative to tourniquet use in CTR. The overall rate of complications was low, and there were no major differences in postoperative outcomes between groups.

Introduction

Carpal tunnel syndrome (CTS) is the most common compressive neuropathy of the upper extremity, with an estimated prevalence of 1% to 5% in the general population.^3,13,27,32 Median nerve compression in the carpal canal impairs epineural blood flow and axonal conduction causing numbness, paresthesias, and weakness in the palmar, radial hand. Wrist splints, anti-inflammatory medications, lifestyle modification, and steroid injections are first-line treatments for CTS; however, a large portion of patients fail to respond to medical management¹⁷ and require surgical release of the transverse carpal ligament for symptomatic relief.

Carpal tunnel release (CTR) can be performed under general or local anesthesia, with or without a tourniquet. Wide-awake hand surgery (local anesthesia, no sedation) is gaining in popularity.² Studies have shown it to be safe, cost-effective, and well tolerated by patients.^1,5,14 Tourniquet discomfort is a downside in wide-awake patients.²² Wide-awake patients can tolerate a forearm tourniquet for 18 to 20 minutes on average,^8,25 and while this is adequate time to complete most CTRs, patients often complain that tourniquet pressure is the most uncomfortable part of the procedure.^4,5,27 Local anesthetic containing epinephrine is an alternative means of achieving hemostasis; it is simple to administer, safe, and effective in CTR.²⁰ The long-standing belief that epinephrine causes digital necrosis has been disproven.^10,21,24,34 The purpose of this study was to evaluate outcomes in wide-awake CTR and to compare tourniquet versus no tourniquet use.

Methods

A database was created of all patients undergoing wide-awake, open CTR by the plastic surgery service at a single institution from February 2013 through April 2016. Data were collected and reviewed retrospectively. Diagnosis of CTS was made clinically and confirmed with nerve conduction studies in 96.7% of cases. All procedures were performed with field sterility in a minor procedure room under local anesthesia (1% lidocaine with 1:100 000 epinephrine). When a tourniquet was used, it was applied to the forearm and inflated to 250 mm Hg just prior to incision. The tourniquet was deflated before skin closure to inspect for hemostasis. Cautery was not routinely used; when necessary, direct pressure was applied to the wound. Postoperatively, all patients were placed in a soft, compressive dressing for 3 days and then routinely seen for follow-up at 2 weeks for wound inspection and suture removal.

Collected variables included age at surgery, medical comorbidities, antithrombotic use and type, current tobacco use, median nerve motor latency at wrist, tourniquet use, operative time, estimated blood loss (as documented in the operative report), complications, and relief of symptoms. Patients were divided into 2 cohorts—tourniquet versus no tourniquet—and outcomes were compared. Cases where multiple procedures were performed or an operative time was not recorded were excluded.

Data were compiled in a worksheet and then transferred to SPSS (SPSS Inc, Chicago, Illinois) for statistical analysis. Patients were divided into 2 cohorts—with and without tourniquet—and evaluated independently. The student t test was used to compare means for continuous data. Fisher exact test was used for categorical data. Multivariate regression was used to control for possible confounding variables, such as sex, presence of medical comorbidities, antithrombotic use, smoking status, and preoperative distal median nerve motor latency. Tests were 2-tailed and results were considered significant for values of p ⩽ .05.

Results

A total of 304 CTRs were performed on 246 patients during the study period. In patients requiring bilateral CTR, the surgeries were performed on separate dates at least 6 weeks apart. Two hundred sixty-nine patients were male (88.5%), and the mean age at surgery was 59.9 years (range, 25.3-91.2 years). One hundred patients (32.9%) were diabetic, and 75 patients (24.7%) were current smokers. Average preoperative median nerve motor latency at the wrist was 6.8 milliseconds (range, 3.0-15.9 ms). There were no major differences in patient sex, age, smoking status, medical comorbidities, or preoperative distal motor latency values when comparing the tourniquet versus no tourniquet groups (Table 1).

Table 1.

Patient Demographics.

	Tourniquet	No tourniquet	P value
Sex, male	81 (90.0%)	188 (87.9%)	.70
Age at surgery (years)	60.55	50.7	.98
Current smoker	19 (21.2%)	57 (26.6%)	.38
Diabetes	32 (35.6%)	69 (32.2%)	.60
Anticoagulated (any type)	34 (37.8%)	65 (30.4%)	.23
Median nerve motor latency at wrist (ms)	6.69	6.78	.81

Ninety-nine patients (32.6%) were on an antithrombotic therapy at the time of surgery. Six patients (4 on warfarin and 2 on dual antiplatelet therapy) temporarily stopped their medication for at least 5 days preoperatively, based on the recommendation of their primary care provider; the remaining 92 patients continued their regimen as usual. There were 69 patients on aspirin, 7 on warfarin, 9 on clopidogrel, 1 on dabigatran, 3 on warfarin plus aspirin, and 4 on dual antiplatelet therapy. Mean operative time was 22.0 minutes for patients on antithrombotics and 24.0 minutes for all other patients (P = .12). Estimated blood loss was 3.73 for patients on antithrombotics and 3.76 for all other patients (P = .92).

A forearm tourniquet was used in 90 of 304 CTRs (29.6%). Mean operative time was 24.97 minutes with a tourniquet and 21.69 minutes without (P = .0029). Estimated blood loss was 3.16 mL with a tourniquet and 4.25 mL without (P < .001). Linear, multivariable regression was performed to control for confounding effects, with operative time as the dependent variable. Female (β = ‒3.71, P = .055), diabetes (β = ‒0.614, P = .653), antithrombotic use (β = ‒3.17, P = .022), and increased preoperative median nerve motor latency at wrist (β = ‒0.424, P = .118) were associated with decreased operative time. Tourniquet use (β = 2.304, P = .082) and smoking (β = 0.222, P = .875) were associated with longer operative times (R² = 0.218, P = .038). The same analysis was performed with estimated blood loss as the dependent variable. Smoking (β = ‒0.102, P = .765), diabetes (β = ‒0.063, P = .859), increased distal median nerve motor latency (β = ‒0.033, P = .614), and tourniquet use (β = ‒1.358, P < .001) were associated with decreased blood loss. Female (β = 0.350, P = .452) and patients on antithrombotics (β = 0.298, P = .386) had higher estimated blood loss (R² = 0.286, P = .005). There was no statistical difference in operative time or blood loss when comparing tourniquet vs no tourniquet use after controlling for confounding variables.

The overall rate of complications was 6.7% with a tourniquet and 5.6% without (P = .79). Surgical site infection (SSI) occurred in 3.3% and 2.8% of patients, respectively (P = .73). Eight out of 9 patients with SSIs were treated successfully with oral antibiotics. One patient required hospital admission and treatment with intravenous antibiotics. Wound healing complications (dehiscence and delayed wound healing) were uncommon in both groups—2.2% versus 2.8% (P = 1.0). No patient required reoperation for any reason during the study period. Overall, 91.8% of patients reported improvement in pain or numbness in follow-up. This value was identical in both the tourniquet and no-tourniquet groups. Mean follow-up time was 3.4 months for those who reported improvement in symptoms and 2.3 months for those without improvement (P = .39) (Table 2). Three attending surgeons staffed all procedures. A total of 29 surgical residents participated in 292 cases (96.1%).

Table 2.

Outcomes.

	Tourniquet	No tourniquet	P value
Operative time (min)	25.52	21.45	<.001
Estimated blood loss (mL)	3.28	4.19	<.001
Complications
Infection	3 (3.3%)	6 (2.8%)	.73
Oral antibiotics	3 (3.3%)	5 (2.3%)	.70
Intravenous antibiotics	0 (0%)	1 (0.47%)	1.00
Admission	0 (0%)	1 (0.47%)	1.00
Wound dehiscence	2 (2.2%)	4 (1.9%)	1.00
Delayed wound healing	0 (0%)	2 (0.93%)	1.00
Other	1 (1.11%)	0 (0%)	1.00
Overall	6 (6.7%)	12 (5.6)%	.79
Outcomes
Improvement of symptoms	78 (91.76%)	156 (91.76%)	1.00
Mean follow-up (months)	2.83	3.10	.69
Symptoms improved	2.96	3.11
No improvement	1.17	2.96

Discussion

Hemostasis in hand surgery has traditionally been achieved by application of a pneumatic tourniquet. Tourniquets create a blood-free operative field which is thought to enable better visualization, facilitate dissection, and improve efficiency. In a 2011 survey of over 700 members of the American Society of Surgery of the Hand, 95% of surgeons reported using a tourniquet for CTR.²³ The trend toward wide-awake hand surgery has brought into question whether tourniquet use is necessary. A patient’s ability to tolerate a tourniquet is often the time-limiting factor in wide-awake hand surgery.

Tourniquets cause pain by direct mechanical pressure and the resultant tissue anoxia. Local risks include soft tissue or neurovascular injury due to cuff compression,^9,11,16,29 chemical burns from skin preparation agents,³⁷ and friction burns from ill-fitting or improperly padded tourniquets.³⁰ Ischemia causes temporary changes in forearm musculature and contributes to postoperative edema.³⁵ Systemic events are rare but include deep vein thrombosis/pulmonary embolism, metabolic and lactic acidosis, electrolyte imbalances, and, rarely, hypotension or myocardial depression secondary to the release of inflammatory modulators.⁹

Local anesthesia with epinephrine is safe and effective in hand surgery; the belief that epinephrine causes digital necrosis has been disproven.^10,21,24,34 A critical review of the literature from 1880 through 2005 by Thomson et al revealed 21 reported cases of fingertip necrosis attributed to the injection of local anesthesia containing epinephrine.³⁴ Most of these cases were from pre-1950 case reports, and all used epinephrine in conjunction with procaine or cocaine, two agents known to cause digital infarction.^7,12,19 To date, no verified case of fingertip necrosis from epinephrine has been reported.²¹ Multiple studies involving thousands of patients support the premise that epinephrine use in digits does not deserve its dangerous reputation.^{6,15,31,33,36}

Previous studies have shown no difference in procedural difficulty or outcome at 6 weeks when comparing tourniquet use with local anesthetic containing epinephrine for CTR.^4,27 This study supports these findings; there was no significant difference in operative time or estimated blood loss with and without a tourniquet after controlling for confounding variables. The overall rate of complications in this study was low and there were no major differences in postoperative outcomes between groups, suggesting that local anesthetic and epinephrine is a safe and effective alternative to tourniquet use in CTR.

Over 91% of patients in this study reported improvement of symptoms (pain, numbness) at the 2-week follow-up appointment. This is in line with success rates of CTR reported in the literature: 70% to 90%.²⁸ Patients who were followed up postoperatively for longer periods were more likely to have improvement in symptoms. This is consistent with previous studies which show an immediate, substantial improvement in pain, numbness, and nighttime awakenings, followed by slow, continued symptomatic improvement as edema and inflammation resolve. Patients with severe CTS, characterized by prolonged distal latencies plus absent sensory nerve action potential or absent thenar compound muscle potential, may continue to improve over the course of a year.¹⁸

There are several limitations to the current study. This is a retrospective review which is inherently prone to patient and treatment selection bias. The study was performed at a single Veterans Affairs medical center. The majority of patients were older males with moderate to severe disease at diagnosis and a high rate of medical comorbidities. Outcomes may differ in a more heterogeneous patient population. Standardized pain scores were not available for all patients. Postoperative improvement in carpal tunnel symptoms was determined based on documentation in the chart during follow-up evaluation; this may cause observer bias. Findings are reported based on the experience of 3 surgeons and may not be the norm for an entire population of surgeons performing CTR at academic institutions. Differences may include case mix, surgical volumes, surgeon seniority, and teaching style, as well as resident participation.

Conclusion

Carpal tunnel release is the most common, nontraumatic hand surgery performed in North America.^3,23 It is typically performed on an elective, outpatient basis using local anesthesia, with or without sedation.²⁶ Poor tourniquet tolerance is a primary concern in wide-awake CTR. Tourniquets create a blood-free operative field but may be of limited value in carpal tunnel surgery due to predictable anatomy and limited required dissection.

In this study, operative time and estimated blood loss were similar with and without a tourniquet. The overall rate of complications was low and there were no major differences in postoperative outcomes between groups. This suggests that local anesthetic with epinephrine alone is a safe and effective alternative to tourniquet use in CTR.