Professional Documents
Culture Documents
Safe Tourniquet Use A Review of The Evidence JAAOS
Safe Tourniquet Use A Review of The Evidence JAAOS
Abstract
Peter G. Fitzgibbons, MD Due in part to an emphasis on quality and cost control within
Christopher DiGiovanni, MD healthcare institutions, protocols for healthcare practice are
increasingly being developed in an effort to maintain normative
Sayed Hares, MD
Downloaded from http://journals.lww.com/jaaos by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/14/2022
Table 1
Published Recommendations on Tourniquet Use
Organization/Study Pressure Duration (min) Reperfusion Interval
Association of Surgical Upper extremity, 50 mm Hg above SBP; Upper extremity, 60; 15 min
Technologists3 lower extremity, 100 mm Hg above SBP lower extremity, 90
Association of periOperative 40 mm Hg above LOP for LOP <130 Upper extremity, 60; 15 min deflation after every 1 h
Registered Nurses1 mm Hg; 60 mm Hg above LOP for LOP lower extremity, 90 of tourniquet time
<131–190 mm Hg; 80 mm Hg above
LOP for LOP >190 mm Hg
Wakai et al2 General recommendation, 50–75 mm Hg 120 30 min at 2-h point in surgery
above LOP; upper extremity, 50–75 lasting >3 h
mm Hg above SBP; lower extremity,
90–150 mm Hg above SBP
Kam et al4 50–150 mm Hg above SBP, using the 120 10 min at the 2-h point for
lower end of the range for the upper surgery lasting >2 h
extremity and the higher end for the
lower extremity
Noordin et al5 Use LOP. No margin specified. 120 NR
important factors include skin pro- Muscle Injury chemic muscle.12 In a rabbit model,
tection under the tourniquet, blood Skeletal muscle injury is reasonably quadriceps musculature compressed
loss, and techniques for attenuating well documented. Studies have mea- at 350 mm Hg of pressure for 2
metabolic changes. sured tourniquet-induced changes in hours demonstrated markedly de-
technetium-99m pyrophosphate creased function at 2 days postopera-
( 99m
Tc PYP) uptake in muscle, con- tively (21% of normal).49 However,
Complications of
tractile function, and histology. In a the muscle demonstrated restoration
Tourniquet Use
rabbit model, 99mTc PYP uptake was to 83% of normal at 3 weeks. Lower
Tourniquet-related pathogenesis is found to be elevated following use of leg muscles experienced a milder dec-
related to ischemia and compression. a thigh tourniquet for 2 hours at 200 rement initially than did the quadri-
The metabolic effects are related to or 350 mm Hg of pressure and for 4 ceps, with variable recovery to near
the sequelae of ischemic tissue, hours at pressure as low as 125 mm normal levels.
whereas nerve and muscle damage Hg.48 Uptake was greatest at the site Multiple studies have documented
are more likely a function of direct of compression in the thigh, which is tourniquet-induced muscle injury at
compression beneath the tourniquet indicative of particular vulnerability a histologic level.11 In a canine study,
itself. Controversy exists regarding to compression. early signs of muscle damage, includ-
the influence of tourniquet design on Tourniquet use impairs muscle ing granular degeneration, inflam-
compressive damage to the soft tis- contractile function, with com- matory reaction, and edema, were
sues. Tourniquet width is of particu- pressed muscle experiencing greater found following 1 to 2 hours of com-
lar interest. dysfunction than the more distal is- pression at 350 mm Hg.6 In a rabbit
Dr. DiGiovanni or an immediate family member has received royalties from Extremity Medical; is a member of a speakers’ bureau or
has made paid presentations on behalf of BioMimetic Therapeutics and Extremity Medical; serves as a paid consultant to or is an
employee of BioMimetic Therapeutics and Extremity Medical; has received research or institutional support from BioMimetic
Therapeutics; has stock or stock options held in BioMimetic Therapeutics and Extremity Medical; and has received nonincome
support (such as equipment or services), commercially derived honoraria, or other non–research-related funding (such as paid travel)
from CuraMedix and Performance Medical Group. Dr. Akelman or an immediate family member has received royalties from Integra
LifeSciences; is a member of a speakers’ bureau or has made paid presentations on behalf of Auxilium Pharmaceuticals; serves as a
paid consultant to or is an employee of BioMimetic Therapeutics; has received research or institutional support from Auxilium
Pharmaceuticals; has stock or stock options held in BioMimetic Therapeutics and OsteoSpring Medical; and serves as a board
member, owner, officer, or committee member of the American Society for Surgery of the Hand. Neither of the following authors nor
any immediate family member has received anything of value from or owns stock in a commercial company or institution related
directly or indirectly to the subject of this article: Dr. Fitzgibbons and Dr. Hares.
Table 2
Findings of Animal Model Tourniquet Studies
Pressure
Study (mm Hg) Duration Findings Model
Table 2 (continued)
Findings of Animal Model Tourniquet Studies
Pressure
Study (mm Hg) Duration Findings Model
ATP = adenosine triphosphate, CPK = creatine phosphokinase, EMG = electromyography, NA = not available, NCV = nerve conduction
velocity
model, notable focal and regional ne- (EMG), and nerve conduction veloc- minutes to complete myelin dissolu-
crosis was noted in the thigh 2 days ity (NCV) studies. Compared with tion and Schwann cell hypertrophy
after a 4-hour tourniquet time at 350 skeletal muscle, nerve tissue seems to at 3 hours. Myelin sheath rupture
mm Hg of pressure.48 Cellular infil- be less vulnerable to acute injury, al- appeared after 2 hours; this finding
tration and milder necrosis were though the effects of injury appear to has been confirmed in other stud-
noted in the leg. Marked but less se- be longer-lasting in nerve tissue fol- ies.10 Similar findings have been
vere changes were seen with tourni- lowing mild to moderate insult. noted in light microscopy studies,
quet pressure measuring 350 mm Hg In one histologic study, rat sciatic with mild changes at 2 hours pro-
for a duration of 2 to 3 hours. nerve structure was evaluated with gressing to significant edema with
electron microscopy 14 days after prolonged times at tourniquet pres-
Nerve Injury application of a tourniquet at 300 sure of 350 mm Hg.50 As with skele-
Nerve injury caused by tourniquet mm Hg of pressure for periods of 30 tal muscle studies, damage was
use has been documented on histo- minutes to 3 hours.8 Damage ranged worse in compressed tissue than in
logic examination, electromyography from no change to mild change at 30 ischemic tissue.
Table 3
Findings of Clinical Tourniquet Studiesa
Pressure
Study (mm Hg) Duration (min) Findings Study Parameters
Arciero et al28 269 (avg) 87 NS EMG changes at 1 mo, NS atrophy at Tourniquet vs none in
1 mo, no differences at 6 or 12 mo ACL reconstruction
Clarke et al33 225 and 300 90 Higher pressure led to greater wound High vs low pressure
hypoxia without clinical complication for TKA
Daniel et al29 Range, 250–300 Range, 40–186 Decreased quadriceps strength and girth Tourniquet vs none in
at 12 wk in the tourniquet group. No ACL reconstruction
significant difference at 52 wk.
Dobner and Nitz34 393 (avg) 42 EMG changes at 6 wk. No EMG changes Clinical
at 5 mo. Posterior tibial nerve most
affected.
Fahmy and Patel30 500 Range, 115–127 Increased fibrinolytic activity returned to Knee arthrotomy
baseline after 30 min
Finsen and SBP + 100 Range, 12–70 No neurologic symptoms at 10 wk Calf tourniquet for
Kasseth35 forefoot surgery
Girardis et al36 350 Range, 75–108 Degree of hemodynamic and metabolic ACL reconstruction
alterations correlates with tourniquet
time
Hirota et al37 300 (avg) Range, 15–150 Amount of emboli in right atrium corre- Knee arthroscopy
lates with tourniquet time. No clinically
significant consequences.
Horlocker et al31 300 145 (avg) Overall risk of neurologic complication, Primary or revision
7.7%. The risk was higher with higher TKA
tourniquet times. Longer reperfusion
intervals were more beneficial. Nearly
all palsies recovered fully.
Kirkley et al38 300 <60 No difference in functional scores at 2, 6, Tourniquet vs none in
and 12 wk. Better visualization with knee arthroscopy
tourniquet up. Increased postoperative
pain with inflation time >30 min.
Kokki et al39 250 101 ± 34 Metabolic changes more significant with ACL reconstruction
longer tourniquet times. No clinical
complications.
350 108 ± 25 Metabolic changes more significant with ACL reconstruction
longer tourniquet times. No clinical
complications.
Konrad et al40 350 56 (mean) Increased pain and swelling in tourniquet Tourniquet vs none in
group at 6 wk. Trend toward decreased ankle ORIF
ROM in tourniquet group.
Lin et al41 480 60–90 Pulmonary functions parameters (Pao2, Tourniquet vs none for
a/A ratio) decrease with tourniquet use. lower extremity sur-
No clinical pulmonary complications. gery
Mittal et al32 SBP + 20–30 15 At 15 min, MNCV was 74% with a wide 14- vs 7-cm–wide cuff
cuff and 83% (avg) with a narrow cuff. on healthy volun-
Wide cuffs have a greater effect on teers
nerve conduction.
Orbay et al42 250–350 100.4 ± 2.4 No complications Narrow silicone ring
stockinette tourni-
quet vs 8-cm pneu-
matic tourniquet
250–350 101.2 ± 26.5 No complications Narrow silicone ring
stockinette tourni-
quet vs 8-cm pneu-
matic tourniquet
Table 3 (continued)
Findings of Clinical Tourniquet Studiesa
Pressure
Study (mm Hg) Duration (min) Findings Study Parameters
Reikerås and 250–350 Range, 78–125 At 4 h postdeflation, D-dimer and pro- TKA
Clementsen43 thrombin fragments are not significant
Rudkin et al44 250 Range, 2–90 Tourniquet time and age are risk factors Foot surgery
for tourniquet pain
Santavirta et al45 450 74 ± 30 Of 1,000 lower extremity cases, 85 ex- Meniscectomy, ankle
ceeded 2 h with no clinical complica- fracture, and TKA
tions reported
Vandenbussche 350 123 No nerve paralysis or wound healing TKA
et al46 difficulty at 3 mo
Weingarden et al47 350–450 53 Avg inflation time of patients with post- Meniscectomy
operative EMG changes, 59 min. Avg
inflation time of patients without post-
operative EMG changes, 41 min. Clin-
ical recovery time increased in pa-
tients with postoperative EMG
changes.
a/A ratio = arterial-alveolar oxygen tension ratio, ACL = anterior cruciate ligament, EMG = electromyography, MNCV = motor nerve conduction
velocity, NS = not statistically significant, ORIF = open reduction and internal fixation, ROM = range of motion, SBP = systolic blood pressure,
TKA = total knee arthroplasty
a
Levels of evidence of the cited studies: level I (28, 34, 38, 40, 41, 46), level II (29, 32, 33, 39, 42, 43, 47), level III (44, 45), level IV (30, 31,
35-37).
Nerve injury has been measured on sue. Lactic acid, pH, glucose, and reac- Coagulopathy and Deep
EMG and NCV studies, as well.9 tive oxygen metabolites such as Vein Thrombosis
One clinical study documented hypoxanthine and xanthine are acutely Several clinical studies have demon-
changes on EMG in 62.5% of ortho- affected by tourniquet use.52-55 strated increased fibrinolytic activity
paedic patients postoperatively.51 Most studies have reported a rapid following tourniquet use, but no
EMG abnormality lasted an average return to normal local acid-base bal- clinically significant changes have
of 51 days (patients were examined ance following tourniquet release. been documented.56-58 Multiple stud-
monthly), and prolonged tourniquet For example, pH has been found to ies have investigated the role of the
time contributed to the incidence and normalize within 20 minutes follow- tourniquet in the production of pul-
severity of abnormalities. NCV stud- ing pressure of 300 mm Hg for 3 monary emboli and the resulting
ies in rabbits have shown significant hours in a canine model16 and within clinically significant cardiopulmo-
changes with compression lasting 2 40 minutes following pressure of 300 nary symptoms. Pulmonary emboli
and 4 hours at tourniquet pressure of mm Hg for 4 hours in monkeys.7 In have been recorded by transesopha-
350 mm Hg, although notable humans, measurements in the right geal echocardiogram during knee
changes are mild or absent in distal atrium have demonstrated minor arthroscopies and arthroplasties per-
nerves.26,27 Clinically, the incidence of changes in pH after tourniquet re- formed with and without tourni-
peroneal and tibial nerve palsy rises lease, and metabolic indicators have quets.37,57,59 In one study, the inci-
with tourniquet times >150 min- been found to normalize within 120 dence ranged from 6% to 79%
utes.31 minutes of tourniquet cessation.7 In depending on the procedure per-
dogs, adenosine triphosphate levels formed, with a nonstatistically sig-
Metabolic Dysfunction do not fall at tourniquet durations of nificant increased incidence of em-
Studies measuring metabolic parame- <3 hours. Although the adenosine boli in the patients on whom a
ters have attempted to address the triphosphate buffer phosphocreati- tourniquet was used.37 None of the
mechanisms underlying tourniquet- nine drops markedly, it is usually re- studies found clinically relevant cases
induced injury of muscle and nerve tis- constituted within minutes.14 of pulmonary embolus. Another
inflation pressures reported an aver- for the lower extremity. No clinical studies have demonstrated clinical
age thigh tourniquet pressure of 338 studies exist indicating that either ex- benefit in terms of neurologic
mm Hg.28-31,33,34,36-41,44-46 In a large tremity is more prone to injury than changes or functional outcomes with
study of tourniquet use and its com- the other. One study measuring the the use of a particular tourniquet de-
plications, the average pressure was pressure at which capillary bleeding sign.
300 mm Hg.37 Animal data clearly occurs found lower values for the
demonstrate that higher inflation upper extremity than for the lower
pressures impart greater insult on extremity.62 A tourniquet pressure of Summary
compressed nerve and muscle than 200 mm Hg in the upper extremity
do lower pressures.10,13,50 Most of Clinical situations involving tourniquet
and 250 mm Hg in the lower ex-
these changes have been shown by use require at least three decisions: the
tremity was found to be adequate to
using tourniquet pressures up to type (ie, shape) of tourniquet, inflation
produce a bloodless field in normo-
1,000 mm Hg, so it is difficult to ex- pressure, and continuous duration of
tensive persons of average build.
trapolate from the literature the sig- occlusive pressure. These questions are
This difference is presumably a func-
nificance of a difference in 25 or 50 often addressed separately in recom-
tion of limb girth, with occlusion oc-
mm Hg at lower pressures. Few clini- mendations. However, studies have
curring at a lower pressure in the up-
cal studies show significant or fre- shown that although each has its own
quent pathology within the ranges per limb. effect on target outcomes, these factors
studied; thus, pressures employed are additive and the exact relationship
clinically (ie, ≤300 mm Hg) seem to Tourniquet Design between them is unclear.9 For exam-
be well within a safe zone of use. The development of new tourniquets ple, the clinical difference between
Several techniques have been de- has spurred debate regarding the 120 and 140 minutes of tourniquet
scribed for determining limb occlu- safety and efficacy of different types time is not clear, and the additional
sion pressure (LOP). Because this of tourniquets. These debates largely impact of 25 or 50 mm Hg more or
measurement accounts for the spe- revolve around the effect of the less inflation pressure on that time
cific tourniquet configuration and width of the tourniquet on both the difference is unknown.
limb girth, it would seem to be a pressure setting required to achieve a The available clinical and basic sci-
more accurate representation of arte- bloodless field and the pathogenesis entific data on tourniquet use do not
rial occlusion than systolic blood of soft-tissue injury. indicate a significant risk of compli-
pressure. Techniques for measuring In a baboon model, Ochoa et al10 cations within the confines of typical
LOP involve the use of a commercial demonstrated nerve damage directly use during orthopaedic surgery, and
device, monitoring of the oblitera- beneath the cuff and suggested that no single standard exists for tourni-
tion of distal pulses with a Doppler the damage is caused by the gradient quet use in all settings. In general,
stethoscope during cuff inflation, or of pressure at the edge of the cuff. for procedures involving <2 hours of
use of a formula based on limb cir- The width of the tourniquet itself tourniquet time, standard practices
cumference.60,61 One report indicates was not studied. However, the impli- should suffice in terms of inflation
that measurement and use of LOP cation that use of a narrower cuff pressure and tourniquet design. Pre-
plus a safety margin may allow for a might result in less nerve damage has operative identification of proce-
tourniquet inflation pressure that is been used in part to justify the recent dures that are likely to involve pro-
lower than the standard fixed infla- development of a narrow nonpneu- longed tourniquet times (ie, >2 hours)
tion pressure.60 Both methods typi- matic silicone ring tourniquet.63,64 allows the surgeon, anesthesiologist,
cally involve a pressure margin A recent study found median nerve and operating room staff ample time to
above the LOP or systolic blood conduction to be more severely affected agree on and institute measures that
pressure that should allow for intra- with the use of a 14-cm tourniquet may ameliorate known risks that are
operative variation in blood pres- than with a 7-cm tourniquet following difficult to clearly define. In such situ-
sure. No studies have demonstrated 15 minutes of inflation.32 With both ations, recommendations include the
a difference in clinical outcomes at- sizes of tourniquet, conduction nor- use of wide, shaped cuffs, the preoper-
tributable to the use of LOP in deter- malized by 30 minutes after defla- ative measurement of LOP by either
mining tourniquet pressure. tion. Other studies have suggested Doppler stethoscope or commercial de-
Recommendations for tourniquet that a wider cuff is safer because it vice, and the use of a reperfusion inter-
pressure setting commonly use lower allows for the occlusion of blood val at 2 hours for procedures lasting
values for the upper extremity than flow at a lower pressure.65,66 No >2.5 hours.
forefoot surgery: Less pain when placed 46. Vandenbussche E, Duranthon LD, 57. Jarrett PM, Ritchie IK, Albadran L, Glen
at the ankle. J Bone Joint Surg Br 1997; Couturier M, Pidhorz L, Augereau B: SK, Bridges AB, Ely M: Do thigh
79(1):99-101. The effect of tourniquet use in total knee tourniquets contribute to the formation
arthroplasty. Int Orthop 2002;26(5): of intra-operative venous emboli? Acta
36. Girardis M, Milesi S, Donato S, et al: 306-309. Orthop Belg 2004;70(3):253-259.
The hemodynamic and metabolic effects
of tourniquet application during knee 47. Weingarden SI, Louis DL, Waylonis GW: 58. Ellis MH, Fredman B, Zohar E, Ifrach
surgery. Anesth Analg 2000;91(3):727- Electromyographic changes in N, Jedeikin R: The effect of tourniquet
731. postmeniscectomy patients: Role of the application, tranexamic acid, and
pneumatic tourniquet. JAMA 1979; desmopressin on the procoagulant and
37. Hirota K, Hashimoto H, Kabara S, et al: 241(12):1248-1250. fibrinolytic systems during total knee
The relationship between pneumatic replacement. J Clin Anesth 2001;13(7):
tourniquet time and the amount of 48. Pedowitz RA, Gershuni DH, Schmidt 509-513.
pulmonary emboli in patients AH, Fridén J, Rydevik BL, Hargens AR:
undergoing knee arthroscopic surgeries. Muscle injury induced beneath and distal 59. Nishiguchi M, Takamura N, Abe Y,
Anesth Analg 2001;93(3):776-780. to a pneumatic tourniquet: A Kono M, Shindo H, Aoyagi K: Pilot
quantitative animal study of effects of study on the use of tourniquet: A risk
38. Kirkley A, Rampersaud R, Griffin S, tourniquet pressure and duration. factor for pulmonary thromboembolism
Amendola A, Litchfield R, Fowler P: J Hand Surg Am 1991;16(4):610-621. after total knee arthroplasty? Thromb
Tourniquet versus no tourniquet use in Res 2005;115(4):271-276.
routine knee arthroscopy: A prospective, 49. Mohler LR, Pedowitz RA, Ohara WM,
double-blind, randomized clinical trial. Oyama BK, Lopez MA, Gershuni DH: 60. Younger AS, McEwen JA, Inkpen K:
Arthroscopy 2000;16(2):121-126. Effects of an antioxidant in a rabbit Wide contoured thigh cuffs and
model of tourniquet-induced skeletal automated limb occlusion measurement
39. Kokki H, Väätäinen U, Miettinen H, muscle ischemia-reperfusion injury. allow lower tourniquet pressures. Clin
Parviainen A, Könönen M, Partanen J: J Surg Res 1996;60(1):23-28. Orthop Relat Res 2004;(428):286-293.
Tourniquet-induced enmg changes in
arthroscopic anterior cruciate ligament 50. Pedowitz RA, Nordborg C, Rosenqvist 61. Tuncali B, Karci A, Tuncali BE, et al: A
reconstruction: A comparison of low and AL, Rydevik BL: Nerve function and new method for estimating arterial
high-pressure tourniquet systems. Ann structure beneath and distal to a occlusion pressure in optimizing
Chir Gynaecol 2000;89(4):313-317. pneumatic tourniquet applied to rabbit pneumatic tourniquet inflation pressure.
hindlimbs. Scand J Plast Reconstr Surg Anesth Analg 2006;102(6):1752-1757.
40. Konrad G, Markmiller M, Lenich A, Hand Surg 1991;25(2):109-120.
Mayr E, Rüter A: Tourniquets may 62. Van Roekel HE, Thurston AJ:
increase postoperative swelling and pain 51. Saunders KC, Louis DL, Weingarden SI, Tourniquet pressure: The effect of limb
after internal fixation of ankle fractures. Waylonis GW: Effect of tourniquet time circumference and systolic blood
Clin Orthop Relat Res 2005;(433):189- on postoperative quadriceps function. pressure. J Hand Surg Br 1985;10(2):
194. Clin Orthop Relat Res 1979;(143):194- 142-144.
199.
41. Lin L, Wang L, Bai Y, et al: Pulmonary 63. Gavriely N: Surgical tourniquets in
gas exchange impairment following 52. Korth U, Merkel G, Fernandez FF, et al: orthopaedics. J Bone Joint Surg Am
tourniquet deflation: A prospective, Tourniquet-induced changes of energy 2010;92(5):1318-1323.
single-blind clinical trial. Orthopedics metabolism in human skeletal muscle
2010;33(6):395. monitored by microdialysis. 64. Drosos GI, Stavropoulos NI, Kazakos K,
Anesthesiology 2000;93(6):1407-1412. Tripsianis G, Ververidis A, Verettas DA:
42. Orbay H, Unlü RE, Kerem M, Sensöz O: Silicone ring versus pneumatic cuff
Clinical experiences with a new 53. Mathru M, Dries DJ, Barnes L, Tonino tourniquet: A comparative quantitative
tourniquet device. Ann Plast Surg 2006; P, Sukhani R, Rooney MW: Tourniquet- study in healthy individuals. Arch
56(6):618-621. induced exsanguination in patients Orthop Trauma Surg 2011;131(4):447-
requiring lower limb surgery: An 454.
43. Reikerås O, Clementsen T: Time course ischemia-reperfusion model of oxidant
of thrombosis and fibrinolysis in total and antioxidant metabolism. 65. Crenshaw AG, Hargens AR, Gershuni
knee arthroplasty with tourniquet Anesthesiology 1996;84(1):14-22. DH, Rydevik B: Wide tourniquet cuffs
application: Local versus systemic more effective at lower inflation
activations. J Thromb Thrombolysis 54. Müller M, Schmid R, Nieszpaur-Los M, pressures. Acta Orthop Scand 1988;
2009;28(4):425-428. et al: Key metabolite kinetics in human 59(4):447-451.
skeletal muscle during ischaemia and
44. Rudkin AK, Rudkin GE, Dracopoulos reperfusion: Measurement by 66. Reilly CW, McEwen JA, Leveille L,
GC: Acceptability of ankle tourniquet microdialysis. Eur J Clin Invest 1995; Perdios A, Mulpuri K: Minimizing
use in midfoot and forefoot surgery: 25(8):601-607. tourniquet pressure in pediatric anterior
Audit of 1000 cases. Foot Ankle Int cruciate ligament reconstructive surgery:
2004;25(11):788-794. 55. Déry R, Pelletier J, Jacques A, Clavet M, A blinded, prospective randomized
Houde JJ: Metabolic changes induced in controlled trial. J Pediatr Orthop 2009;
45. Santavirta S, Kauste A, Rindell K: the limb during tourniquet ischaemia. 29(3):275-280.
Tourniquet ischaemia: Clinical and Can Anaesth Soc J 1965;12(4):367-378.
biochemical observations. Ann Chir
Gynaecol 1978;67(6):210-213. 56. Klenerman L, Chakrabarti R, Mackie I,
Brozovic M, Stirling Y: Changes in
haemostatic system after application of a
tourniquet. Lancet 1977;1(8019):970-
972.