| Sign In to gain access to subscriptions and/or personal tools. |
DOI: 10.1016/J.JHSB.2003.12.008
Stopping Warfarin Therapy is Unnecessary for Hand SurgeryFrom the Department of Plastic Surgery, The Queen Victoria Hospital, East Grinstead, West Sussex, UK Correspondence: Mr H.J.C.R. Belcher, Department of Plastic Surgery, The Queen Victoria Hospital, Holtye Road, East Grinstead, West Sussex, RH19 3DZ, UK. Tel.: +44-1342-414236; fax: +44-1342-414121; E-mail: hjcrb{at}pncl.co.uk
Interruption of appropriate therapeutic warfarin therapy imposes a risk of morbidity and mortality on the patient. Strategies to reduce the risks of interruption impose relatively large costs in terms of prolonged hospital stay, medication and coagulation monitoring. We report a series of 47 consecutive surgical episodes on the hands of 39 patients without interruption of therapeutic warfarin anticoagulation and with an INR of between 1.3 and 2.9. There was no difficulty with intraoperative haemostasis. Two patients had minor bleeding-related complications with no long-term sequelae. The authors conclude that interruption to warfarin therapy is unnecessary if the INR is less than 3.0 and therefore inappropriate for therapeutically anticoagulated patients undergoing hand surgery.
Key Words: warfarin • hand surgery • anticoagulants
Interruption of anticoagulation in patients warfarinized for atrial fibrillation (AF), mechanical heart valves or previous thromboembolism poses a significant risk of morbidity and mortality (British Thoracic Society, 1992; Kearon and Hirsh, 1997). There is a significantly greater risk of a thromboembolic event when oral anticoagulation is manipulated, compared to the background risk without oral anticoagulation (Blacker et al., 2003). These risks may be further exacerbated by the hypercoagulable state that can accompany surgery (Kearon and Hirsh, 1997; Seyfer et al., 1981). Bleeding problems can occur both intraoperatively and postoperatively. Therefore anticoagulative therapy may not be restarted until the perceived risk has significantly decreased (Haemostasis and Thrombosis Task Force of the British Society of Haematology, 1998; Parkin and Manners, 2000; Wahl, 1998). In addition to the loss of thromboembolic protection caused by manipulation of the anticoagulation, there is the significant postoperative problem of restoring the therapeutic INR. This usually entails an inpatient stay, with the associated costs and morbidity. Despite these risks and costs, many authors advocate peri-operative interruption of warfarin therapy (Haemostasis and Thrombosis Task Force of the British Society of Haematology, 1998; Kearon and Hirsh, 1997; Wahl, 2000). Uninterrupted anticoagulation appears to be safe in cataract surgery (Hall et al., 1988; Morris and Elder, 2000), dental (Wahl, 2000) and skin surgery (Alcalay, 2001; Otley et al., 1996), but there are no studies regarding hand surgery (Dunn and Turpie, 2003).
Since 1996 the senior author (HJCRB) has performed hand surgery without interrupting the anticoagulation regimen if the INR was less than 3.0. The medical records of all patients on warfarin treatment who underwent surgery between May 1996 and March 2003 were reviewed. All patients had their INR checked preoperatively. If the INR was greater than 3.0, then procedures were deferred and adjustment was undertaken to bring it into the range 2.0 to 3.0. The INR was later rechecked to confirm the level was below 3.0. The standard surgical management in all cases comprised use of a tourniquet that was applied at the mid-arm level with an inflation pressure of 280 mmHg. The tourniquet was deflated and haemostasis was secured prior to skin closure with an absorbable suture (5-0 Vicryl rapide). Postoperatively, patients were reviewed prior to discharge and were given both verbal and written instructions in their own care. All patients were seen in out-patients at 2 weeks for a wound check, and review. Further follow-up continued as directed by the type of procedure.
There were 55 procedures carried out as 47 surgical episodes on 39 patients. The average INR at the time of surgery was 2.2 (range 1.3–2.9). The indications for anticoagulation are shown in Table 1 and the surgical procedures performed are shown in Table 2. There were 19 men and 20 women. The average age at surgery was 49 (range 31–87) years. Eleven patients were treated as day-cases, 18 patients stayed 1 night, five patients stayed 2 nights and seven patients were either day-cases or stayed one or two nights (incomplete data). Two patients stayed 4 nights and one patient stayed 7 nights for intensive physiotherapy. Three elderly patients stayed 15 (two patients) and 19 nights after traumatic injuries that required skin grafting and a prolonged stay for social reasons. There were no additional days of stay related to bleeding complications.
In all cases intraoperative haemostasis was achieved without difficulty. Two patients in this series developed bleeding-related complications.
Case 1
Case 2
Anticoagulative treatment is necessary to reduce morbidity and mortality in a variety of patients. In the absence of anticoagulation therapy the risk of recurrence in the first 3 months following a proximal deep vein thrombosis (DVT) is approximately 50% (Levine et al., 1995). Of these recurrences, 6% cause death and 2% will result in major morbidity (Levine et al., 1995; Schulman et al., 1995). The risk of arterial embolus in patients with atrial fibrillation varies from 1% to 20% (average 4.5%) per year depending on multiple risk factors (European Atrial Fibrillation Trial Study Group, 1993). Mechanical heart valves pose an annual risk of arterial embolization of approximately 8% (Mok et al., 1985). Arterial thromboemboli cause death in 20% of cases and permanent disability in approximately 40% (European Atrial Fibrillation Trial Study Group, 1993; Kearon and Hirsh, 1997). Day surgery is becoming more common in all branches of surgery. The current consensus recommendations (British Thoracic Society, 1992; Haemostasis and Thrombosis Task Force of the British Society of Haematology, 1998; Kearon and Hirsh, 1997) are to omit warfarin for several days prior to surgery. Unfractionated (intravenous) heparin is recommended to provide thromboembolic protection. Entailing close and frequent monitoring during this period, as an inpatient, with the attendant morbidity and cost. The need for the close monitoring underlies the common problem of sub-therapeutic treatment on starting intravenous heparin (Turpie et al., 2002). On average it takes 4 days of warfarin omission for the INR to reach 1.5. Postoperatively it takes an average of 3 days, with a wide variation, for the INR to reach 2.0 (White et al., 1995). Therefore there is an inevitable time of lower thromboembolic protection, which poses a significant risk of a thromboembolic event (Blacker et al., 2003). Several studies have demonstrated that low molecular weight heparin, given by once daily subcutaneous injections, has superior efficacy, a safer profile and greater cost efficiency than unfractionated (intravenous) heparin for deep vein thrombosis treatment (O’Shaughnessy et al., 1998; Rose et al., 2001; Turpie et al., 2002). Low molecular weight heparin, at treatment doses, has a 24 hour efficacy, which gives it great flexibility for outpatient use but is difficult around surgery. The ability to quickly and reliably reverse the effect of unfractionated heparin is considered important for peri-operative use, and low molecular weight heparin has not been evaluated for peri-operative use. Our protocol of checking the INR preoperatively is part of our current routine when the patient is reviewed in the preadmission clinic. The only extra cost incurred is when the INR is greater than 3.0 and requires adjustment, and a recheck. This is clearly cheaper than an inpatient stay for unfractionated heparin or low molecular weight heparin administration under outpatient nurse monitoring, if licensed. There is great value in the protocol being clear and simple to follow for all the members of the surgical team. In our study, we noted no problems with haemostasis intraoperatively. This is consistent with other evidence that demonstrates that the surgeon is unable to predict the INR from the intraoperative findings (West et al., 2002). Furthermore, the bleeding time does not relate to difficulty with haemostasis (Lind, 1991). In ocular surgery no difference was found between patients who had stopped, and those who had continued, their warfarin (Gainey et al., 1989). In this study of a diverse population of 47 anticoagulated patients we had a bleeding problem or haematoma rate of 4%. Due to the small numbers of specific conditions one cannot attempt a valid comparison to the published rates of complications, but our experience is comparable to the complication rates for several studies of hand surgery relating to Dupuytren’s contracture (Table 3). Therefore we are able to demonstrate good safety across a broad selection of procedures.
We recommend that an INR of less than 3.0 is a safe value for hand surgery. We have found that the described protocol is simple for the surgical team to follow and avoids the increased morbidity and mortality rate associated with manipulation of therapeutic anticoagulation.
Received for publication May 29, 2003. Accepted for publication December 18, 2003.
  CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
Abstract
INTRODUCTION