INTRODUCTION & HISTORY
COMPLICATIONS & PROBLEMS WITH USE
The purpose of this presentation is to discuss the arterial tourniquet, considering all it's uses under various types of anesthesia both regional and general and for intravenous regional anesthesia (IVRA). Most of the recommendations are rather conservative and represent suggestions of various authors rather than a "standard" for practice. The arterial tourniquet is unphysiologic and associated with a number of problems.
Arterial Tourniquet Uses :
Positions Used :
Use of the tourniquet anywhere but on the proximal limbs (upper arm and thigh) is often discouraged because of the increased risk of complications, particularly neurologic injury with distal tourniquets. Responsibility for the tourniquet - maintenance, application, and potential complications falls on both the anesthesiologist and the surgeon. Legally it may be analogous to positioning; all are responsible (whether the tourniquet is used primarily for anesthetic or surgical indications).
The Esmarch was the first tourniquet used. A rubber band was wrapped around the limb several times. The primary problem was the high pressures generated. Pressures in excess of 1000 mmHg have been demonstrated. As well, twisting and stretching of skin during application promoted skin trauma. These problems produced a higher incidence of complications associated with the Esmarch tourniquet relative to the pneumatic tourniquet. There was definitely a greater occurrence of nerve injury during early use as compared to the pneumatic tourniquet. As a result, the Esmarch has generally been abandoned.
The tourniquet must be used under the direct supervision of those experienced with it and responsible for it .
Most tourniquet-related morbidity occurs as a result of equipment malfunction ! Therefore, scheduled maintenance is important. Tourniquets should be of non-slip orthopedic type for all applications including IVRA. As wide a cuff as possible should be used. The entire apparatus should be checked, including - pressure source, pressure gauge, regulator, tubing, connectors, inflatable cuff and casing. A common problem relates to faulty pressure gauges. Once the entire apparatus is inspected and tested, a note should be written as to it's working condition prior to use. The use of automatic-gas operated tourniquets has been discouraged since their use was one of the common factors in the seven reported fatalities secondary to IVRA. (Heath 1982, 1983 / Britain Department of Health and Social Security, Hazard Notice, 1982)
Exsanguination of the limb prior to tourniquet inflation decreases the amount of blood distal to the cuff. This reduces blood in the surgical field and may limit peak plasma levels of local anesthetic (LA) in the case of IVRA when the cuff is deflated.
Relative Contraindications to Limb Exsanguination:
No standards exist, however, many recommendations have been proposed. The tourniquet must be applied by someone experienced in it's use, function, and complications. When used for IVRA, it should be applied by the anesthetist responsible. Padding is applied first, smooth and snug, to prevent skin trauma. Prior to inflation, all air should be removed from the cuff and the made to fit smoothly without wrinkles. It is applied closely and snugly to the padding to avoid skin trauma. In obese patients, draw the skin and subcutaneous tissue distally during application. Helps to prevent distal slippage with subsequent loss of occlusion and potential LA toxicity (IVRA) and loss of hemostasis. Proximal limb placement (thigh or upper arm) is preferred because larger amounts of tissue and muscle protect nerves from potential trauma. Place the tourniquet on the point of maximal circumference of the limb. Placement below the elbow or knee invites a higher risk of complications and is discouraged by some authors. Limited information is available comparing distal to proximal placement and incidence of complications. Avoid application over bony prominence or any area where skin, nerves, or blood vessels are compressed excessively against a hard bony surface. Beware bony deformities and previous fracture sites. There are 2 case reports of radial nerve paralysis with tourniquet use in patients with a past history of humeral fracture. (Ann Plas Surg, 1990;24;4;346).
The tourniquet should be inflated rapidly to prevent blood from being trapped in the limb during the period when the cuff pressure exceeds venous but not arterial pressure.
Effects of Tourniquet Inflation :
The aim is to produce a bloodless field and/or containment of local anesthetic. Therefore, the cuff pressure needs to be high enough to prevent arterial and venous blood from passing beneath. However, the major mechanism of nerve injury related to the tourniquet is excessive pressure causing nerve compression. Therefore, a compromise has to be made. The lowest pressure that safely assures hemostasis and arterial occlusion is the goal.
This minimum tourniquet pressure required to maintain hemostasis will depend on :
Safe Approach = Measure the patient's SBP preoperatively and look at chart to note usual BP range. Inflate the tourniquet an additional 50 to 75 mmHg for the arm and 75 to 100 mmHg for the leg above the baseline SBP. The technique has been shown effective in a prospective study of 84 patients. (Clin Orhto Rel Res, 1983;177;230) Never clamp tubing to prevent deflation. A leak in the cuff itself may not be detected if the pressure line is clamped.
Minimum - Applies to IVRA. 15 to 25 minutes have been recommended. The goal is to minimize systemic toxicity. With 0.5% lidocaine 2.5mg/kg in the arm and tourniquet deflation after 5 minutes, peak venous levelis < 2mcg/ml.
Maximum - Applies to any case. Absolute limits for tourniquet ischemia and nerve compression have not been established or agreed upon. There is no rule as to how long a tourniquet can be inflated safely. Maximum safe duration recommended on literature review ranges from 1 to 3 hours, most commonly 1.5 to 2 hours.
Shaw-Wilgis - Biochemical (Venous pH, pO2, pCO2 in the arm) (J Bone Joint Surg. 1971 ; 53-A ; 1343)
n=50 Value pH pO2 pCO2 Preinflation 7.40 45 68 1 Hour 7.19 20 62 2 Hours 6.9 4 164
Therefore, the recommended maximum time is 2 hours. pH of 6.9 corresponding to the fatigue point of muscle and further decreases may produce irreversible damage leading to postop muscle weakness. Histologic studies generally show early changes at > 1 hour but muscle degeneration and cell necrosis occurs at 2 to 3 hours. Functional studies show that most patients tolerate 2 hours of tourniquet ischemia with no sequelae. However, tourniquet paralysis has been reported when tourniquet time and pressure have been "safe". As well, EMG abnormalities and more subtle functional changes have been demonstrated with times < 1 hour.
Other factors are considered in determining safe tourniquet time, including the patient's general health, nutritional status, tourniquet pressure, limb trauma, neuropathy, PVD, etc. These factors alter a patient's susceptibility to complications and can influence decisions about safe duration.
There is NO SAFE MAXIMUM TOURNIQUET TIME ! The safest time is the shortest time. Current information suggests continuous application should not exceed 2 hours.
If prolonged tourniquet times are necessary, then reperfusion periods should be used. This allows correction of metabolic abnormalities in the limb and restoration of depleted energy stores. Studies suggest reperfusion times of 5 to 20 minutes with most recommending 15 to 20 minutes after an initial 2 hour application. Optimal timing of reperfusion for subsequent ischemic periods is unknown.
Systemic effects of Tourniquet Deflation :
Changes can be reduced with short tourniquet times, ETCO2 monitoring, and controlled ventilation. Preexisting metabolic abnormalities and underlying disease may make an arterial line advisable in the at-risk patient. Monitoring of pulse, BP, ECG, respiratory status, and neuro status are important when the tourniquet is being deflated and during the period of recovery from the above abnormalities.
FACTOR CHANGE pO2 Mild Decrease SaO2 Little Change SvO2 Mild-moderate Decrease pH Mild Decrease pCO2 Mild Increase ETCO2 Mild Increase Lactate Increase K+ Increase Core Temp Decrease CVP Decrease SBP Decrease
Method for IVRA :
The goal here is to minimize bolus drug release and decrease the peak concentration achieved. Keeping plasma levels as low as possible helps avoid toxic side-effects. Factors which may affect peak local anesthetic level are :
When the arterial tourniquet should be avoided. Many are of obvious concern.
Sickle Cell Disease
Use in these patients is controversial. Remember, sickling is promoted by hypoxemia, acidosis, and circulatory stasis.Therefore, a tourniquet places the patient at risk. Systemic metabolic abnormalities induced by tourniquet deflation can potentially produce sickling in patients with Sickle cell anemia (HbSS) and Sickle trait (HbAS). Despite the theoretical contraindications, available data does not reveal an increased incidence of problems in people with HbSS or HbAS. However, these reports are few and retrospective. Therefore, the use of arterial tourniquets in patients with HbSS or HbAS cannot be recommended pending further studies. If tourniquet use is absolutely necessary, certain steps can be taken :
Complications occur with improper and proper use of the tourniquet. Complications can be associated with all phases of tourniquet use - exsanguination of the limb, tourniquet inflation, maintenance of tourniquet inflation, failure to maintain tourniquet inflation (failure), and tourniquet deflation.
Exsanguination of a limb autotransfuses blood from the peripheral circulation into the central circulation.
Bradford (Anesthesia, 1969;24;2;190)
- Exsanguination/tourniquet on one leg prouces a mean CVP increase of 9.7 cmH2O
- Exsanguination/tourniquet on two legs produces a mean CVP rise of 14.5 cmH2O maintained until tourniquet release in 80%. Estimated exsanguination of both legs adds 700-800 ml to the central circulation.
When cardiac reserve is poor, this volume load may not be tolerated. Cardiac arrest following bilateral leg exsanguination has been reported. If signs of volume overload develop, immediate deflation is necessary. At that point, tourniquet use is abandoned or used with appropriate fluid management, vasodilation, and monitoring.
There are 2 case reports of fatal pulmonary emboli secondary to limb exsanguination and tourniquet inflation. Both patients were hospitalized > 2 weeks with lower extremity injuries. (Anesth Analg, 1984;63:371) Preop evaluation for venous thrombosis is therefore important, especially when risk factors for venous thrombosis are present, such as immobilization > 3 days and trauma. Positive findings are relative contraindications to exsanguination and arterial tourniquet. A high index of suspicion is necessary as DVT's may be clinically silent.
Pulmonary embolism has also been associated with tourniquet deflation. There are 2 case reports of pulmonary embolism following tourniquet deflation in knee surgery. These patients had no preop evidence of DVT. In one case preop noninvasive vascular studies failed to detect DVT. In both cases, < 5 minutes after deflation, the patients developed severe cardiorespiratory distress resulting in cardiac arrest. Both were resuscitated. The patient with negative preop studies was placed on fem-fem bypass and had a median sternotomy in preparation for open pulmonary embolectomy. In this case, the diagnosis of PE was made with TEE. This patient died 3 days later, the other patient survived.
(Anesthesiology, 1991, 74;3;618-20)
Placement of Padding & Cuff:
Trauma secondary to improper placement of padding or the tourniquet cuff can occur. Loose or wrinkled application of either can lead to skin pressure trauma as the cuff is inflated. Abrasions, bruising, ecchymosis, and blistering can result and have resulted in legal action.
Skin burns can occur when prep solutions get under the cuff. Full thickness burns have been reported (Aust NZ J Surg, 1978, 48;66). Plastic drape around distal tourniquet prior to prep can avoid this.
Inadequate Hemostasis (4 reasons) :
If bleeding occurs, first check tourniquet function and the patient's blood pressure before making any adjustments. The problems may be with the patient and not the tourniquet. Blindly increasing tourniquet pressure may be of no benefit and can increase the potential for damage.
Local Anesthetic Toxicity during IVRA. Can occur during tourniquet maintenance secondary to venous leak or complete loss of pressure (see below).
Edema/Engorgement can occur secondary to arterial inflow with no venous exit.
Previously discussed - decreased pH, decreased pO2, increased pCO2, increased K+, increased Lactate, etc. Generally these changes are mild and well tolerated. However, one concern that can occassionally be of significance is the influence of alterations in CO2 on CBF. Increases in PaCO2 can lead to increases in CBF and this has implications for the use of tourniquets in head-injured patients. In a study using transcranial doppler to measure MCA flow velocity; increases of 58% and 13% occurred within 4 and 1 minutes after tourniquet release. The flow remained elevated for > 7 minutes. This suggests that significant increases in CBF can occur after intraoperative tourniquet release. This increase seems to be mostly CO2 dependent. (CJA, 1990;37;4;S29)
Tourniquet pain develops in up to 66% of patients, 30 to 60 minutes after cuff inflation in patients receiving regional anesthesia to the arm or leg. Most studies involve spinal or epidural anesthetics. It is described as dull, burning, deep, poorly localized, and aching. It increases steadily until it becomes unbearable. It is relieved immediately on deflation of the tourniquet. It occurs despite adequate anesthesia for surgery, eg. spinal, epidural, arm blocks, etc. It remains one of the main factors limiting tourniquet time in regional anesthesia.
There are a lot of theoretical explanations. However, the bottom line is that the mechanism is not completely understood. No concrete data exist but there are many observations and several proposed pathways. Most types of nerve fibers have been implicated at one time or another.
Painful impulses may travel with unmyelinated sympathetic (C) fibers above level of spinal block. Inadequate block of large nerve fibers may allow compression and ischemia to penetrate block. Possible Pathway = Unmyelinated sympathetic (C) fibers that can travel up the sympathetic trunks and enter the cord above the level of a spinal block or remain unblocked. However, stellates do not block tourniquet pain in the arm. Therefore, sympathetics are likely not involved or sympathetics which are not blocked with stellate may be involved. Possibly secondary to inadequate block of large nerve fibers because of inadequate concentration/dose of LA. Compression and ischemia of large nerves may penetrate the block and produce pain. This is supported by the observation that higher LA doses producing the same spinal level decrease the incidence of tourniquet pain. For example, increasing tetracaine dose from 12 mg to 16 mg produced similar levels of anesthesia and decreased tourniquet pain from 64% to 33%. (Anesthesiology 1962;23;287) The postulate was that a higher concentration of LA was necessary to block large pressure-pain fibers. Tourniquet pain, therefore may have more to do with intensity of block rather than block level (tourniquet pain appears unrelated to spinal level).
Methods to deal with tourniquet pain:
Many different modalities of treatment have been proposed. Therefore, it is unlikely that any one is completely effective.
Tourniquet-induced hypertension occurs in 11-66% of cases. It's onset is analogous to the onset of tourniquet pain, approximately 30-60 minutes. The etiology is unclear. It likely has the same origin as tourniquet pain, and requires a specific critical level of cellular ischemia in muscle or nerve. It is more common with general anesthesia than with regional anesthesia. There is a very low incidence with spinal anesthesia. Sympathectomy does not block it's occurrence. Regional anesthesia may lessen the occurrence of tourniquet hypertension at the expense of a similar incidence of tourniquet pain.
Exsanguination - Volume shift with increase in SVR (SBP) and CVP. CHF and arrest have been reported.
Inflation / Maintenance - Tourniquet hypertension after 30 to 60 minutes.
Deflation - Mild to moderate hemodynamic changes. Transient fall in central venous and systemic arterial pressures as mentioned above. Usually benign, but can be significant in patients with coexisting cardiovascular disease. Fatal myocardial infarction and cardiac arrest have been reported coincident with tourniquet deflation. Factors include sudden reduction in SVR, acute blood loss, and release of ischemic metabolites (eg.thromboxane).
Implications: Must be prepared to deal with hemodynamic changes. Consider invasive monitoring if cardiovascular status warrants.
Arterial thrombosis can result from dislodgement of arterial plaque. Be aware of this problem in patients at risk for arterial occlusion (history of arterial emboli, older, PVD, drugs, carcinoma, etc).
Muscle is more susceptible to ischemic damage than nerve. Muscle injury is more severe under the tourniquet where it is exposed to ischemia and pressure. As tourniquet time increases there is progressive cellular hypoxia, acidosis, and cooling in the occluded limb. Histologic evidence of muscle damage is evident at 30-60 minutes tourniquet time. 2-3 Hours of ischemia produce cellular necrosis and endothelial capillary leak. These changes progress and peak 24 hours after tourniquet release. After tourniquet release, CK and myoglobin levels increase. Significant elevations do not occur unless 2 hours of tourniquet time is exceeded. Physiologic function is abnormal posttourniquet. The muscles ability to develop tension can be decreased for days.
Edema: With deflation there is an immediate swelling of the involved limb. This does not correlate with tourniquet time or pressure. Half is secondary to return of exsanguinated blood and half secondary to post-ischemic reactive hyperemia. Heparin and steroids have not helped. Hypothermia of the operative limb has worked but is difficult to implement. Cold saline irrigation plus quick removal of the cuff and padding at the end help to decrease edema and venous congestion.
Compartment Syndrome: Very rare. Include this in the differential diagnosis of posttourniquet neurologic and vascular dysfunction.
" Posttourniquet Syndrome ": Swollen, stiff, pale limb with weakness but no paralysis. Duration 1-6 weeks. Postoperative edema is the main etiology.
Bleeders may not be identified intraoperatively, therefore closure of the operative site prior to tourniquet deflation can predispose to postoperative hematoma formation, and/or bleeding. Potential for acute blood loss combined with hemodynamic changes from tourniquet deflation and anesthesia should be kept in mind.
Delayed return of blood flow after tourniquet release :
Midazolam study (Anaesthetist 1991;40;2 p79)
Midazolam administration to 3 groups: 1) No tourniquet 2) Midazolam before tourniquet 3) Midazolam after tourniquet.
Fentanyl study - Acta-Anaesthesiol-Scand 1990;34;2
Neurolept anesthesia for total knee replacement. Fentanyl and midazolam were given before tourniquet application. Plasma levels were measured after release.
LA Systemic Toxicity :
Seven deaths were reported from 1979 to 1983. Common factors in these cases included the following:
Systemic side-effects are primarily neurologic with incidence ranging from 2.1% to 67.3% depending on the drug and dose used. There is a higher incidence of side-effects with larger doses and with bupivacaine and lidocaine as opposed to prilocaine. Cardiovascular effects such as hypotension and bradycardia are rare and when they occur are mild and transient. Methemoglobinemia can occur when prilocaine is used. However, this is not usually seen until a dose of > 600 mg is given. Maximum recommended dose for IVRA is 3 to 4 mg/kg.
LA Access to Central Circulation:
Before tourniquet release drug can enter the systemic circulation 3 ways :
After tourniquet release the drug is released in a biphasic manner, with an initial rapid release followed by slow washout. 30% Bolus release occurs within 30 seconds of deflation followed by slower washout (50% of dose remaining at 30 minutes). Factors mentioned previously can affect peak levels (exsanguination, time to tourniquet release, method of release, arm movement after release)
There is a wide spectrum of reported neurologic injuries from paresthesias to complete paralysis. The overall incidence of severe dysfunction is extremely low but the incidence of more subtle neurologic dysfunction is probably much larger than we appreciate. True incidence of specific types of neurologic injury attributable to tourniquet use (paresthesia, paralysis, duration, etc) cannot be determined due to a lack of information in available studies.
4 SIGNIFICANT REVIEWS:
1. Australian Orthopedic Association - Aust NZ J Surg,
2. Flatt, 1972 - Arch Surg, 1972;104;190
3. Lundborg & Rydevik, 1983 - Acta Ortho Scand, 1983;54;669
4. Sherman et al., 1986 - J Bone Joint Surg, 1986;68-A;256
Summary of reports from of tourniquet paralysis from 1931 reveals - 129 cases :
The radial nerve was the nerve most commonly injured in the upper extremity. This is attributed to it's anatomy (closeness of the nerve to bone with little intervening tissue). Tourniquet paralysis in the lower extremity secondary to pneumatic tourniquet is exceedingly rare. Of the 5 cases reported, all had initial pressures set at 500 mmHg. Protection is afforded by the increased muscle bulk relative to the arm and the absence of the nerve close to bone situation (except for common peroneal). Despite the very low occurrence of major neurologic sequelae (ie paralysis) the incidence of more subtle neurologic abnormalities and prolonged functional impairment appears to be quite common.
1. Weingarten & Saunders - EMG studies (JAMA, 1979;241;1248) (Clin Ortho Rel Res 1979;143;194)
2. Dobner & Nitz - EMG and Function (Am J Sports Med, 1982;10;211)
6 weeks postop :
Variable EMG Abnormal Functional Capacity Mean Tourniquet Relative to Normal Leg Time (Single leg vertical leap, % of normal) ------------------------------------------------------------------------------------ Tourniquet 71% 39% 44 minutes No Tourniquet 0% 79% 41 minutes
Other sequelae include stiffness and weakness. Often the more subtle effects are not diagnosed because of dressings, casts, and immobilization. It is believed that the incidence of subtle neurologic injury is more common than previously thought.
The following points reflect the current thinking :
1954 Mouldaver, "Posttourniquet syndrome" :
Natural History of Nerve Injury.
Types of nerve injury :
EMG studies can predict outcome. Conduction block at the tourniquet level with normal conduction below is a usual finding and indicates short term recovery. Conduction block below the tourniquet level predicts a slower recovery.
Tourniquet Associated Nerve Injury :
When it Happens ?
When patients develop neurologic deficits postoperatively in a case where a tourniquet was used, the first thing to remember is that there are many potential causes. Regional anesthetics can be blamed for postop neuropraxias when tourniquets have been used.
Differential Diagnosis of Posttourniquet Neurological Dysfunction:
Because of the potential complications and adverse effects, use of the tourniquet should be kept in mind during the preoperative evaluation. A complete evaluation is necessary prior to any tourniquet use.
Systems Implications for Tourniquet use:
Premed with sedative to increase cuff tolerance if the procedure is expected to last > 30 minutes with a regional anesthetic. Always check the tourniquet as part of the equipment check. Use only proper pneumatic tourniquets, avoid rubber band type (Esmarch) tourniquets. Tourniquet size should be appropriate for arm size, wider cuffs minimize pressure. Larger better. Use lowest tourniquet pressure that provides adequate hemostasis safely. Do not clamp tubing. Continuously monitor tourniquet pressure. Beware Esmarch use in patients at high risk for DVT and PE. Carefully watch volume status in patients with limited cardiac reserve when Esmarching. Apply tourniquet where the nerves are best protected by muscles. Proximal placement avoiding points where nerves are easily compressed is best. Be prepared to treat hemodynamic changes associated with tourniquet inflation, maintenance, and deflation. If the acid / base status of the patient is otherwise compromised, then an arterial line for monitoring may be useful when a tourniquet is used. Shortest possible tourniquet time is important for many reasons (pain, hypertension, trauma, metabolic changes, etc.). Time can be decreased by delaying inflation until the latest possible point. Once the tourniquet has been deflated, immediate removal of the cuff and padding helps decrease edema secondary to venous obstruction after arterial inflow resumes. When prolonged tourniquet times are necessary use a double tourniquet and intermittent reperfusion periods. Do not allow prep solution getting under the cuff.
Department of Anesthesia, Ottawa General Hospital