Spinal and Epidural Blockade and Perioperative Low Molecular Weight Heparin

Smooth Sailing on the Titanic

The R.M.S. Titanic set sail on April 10, 1912. The ship was considered invincible, yet 5 days into her maiden voyage, on a moonlit night with calm seas, the Titanic collided with an iceberg and sank. Although the collision may have been unavoidable, the consequences need not have been catastrophic. Many critical incidents occurred before the ship's sinking and the tragic loss of life, including overconfidence and lack of vigilance by the captain and crew, the presence of too few life boats, and delay in passenger evacuation. Correction of any of these critical incidents may have improved (or possibly prevented) the disastrous outcome of this historic event.

Enoxaparin, the first low molecular weight heparin (LMWH) approved by the Food and Drug Administration (FDA) in the United States, was distributed for general use in May 1993. The approved dose scheduling of enoxaparin was 30 mg every 12 h, with the first dose administered as soon as possible after surgery. Within 1 yr, two cases of spinal hematoma had been voluntarily reported. The warnings section of the drug label was revised in March 1995 to caution practitioners of the risk of spinal hematoma in patients with indwelling catheters or concomitant treatment with antiplatelet medications. In addition, the prescribing information was changed to recommend that the first dose be given 12-24 h after surgery (rather than immediately postoperatively). By October 1995, 11 cases of spinal hematoma had been reported. A second revision of the drug label with expanded warning and adverse reaction sections was requested in January 1996, and a "Dear Doctor" letter was issued by the manufacturer. However, despite efforts at relabeling and education, cases of spinal hematoma continued to occur. A total of 30 cases of spinal hematoma in patients undergoing spinal or epidural anesthesia while receiving LMWH perioperatively were voluntarily reported between May 1993 and November 1997. An FDA Health Advisory was issued in December 1997. In addition, the manufacturers of all LMWH and heparinoids have been requested to revise the labeling of their respective products, specifically the placement of a "boxed warning."

It is difficult to ascertain the actual frequency of spinal hematoma in this patient population. Spinal hematoma is a catastrophic complication; there are undoubtedly unreported cases. Conversely, although the number of doses of LMWH distributed is known, not all were administered to patients undergoing regional anesthetic techniques, and there are many nonsurgical, "off-label" uses of the drugs. Despite these uncertainties, calculations place the frequency of spinal hematoma between 1:1000 and 1:10,000 neuraxial blocks ^[1]. This incidence is much more frequent than that of previous assessments, which have estimated the probability of significant spinal bleeding after spinal or epidural anesthesia to be approximately 1:200,000. Because this represents the upper 95% confidence interval, the actual incidence of spinal hematoma should be much lower, approximately 1:1,000,000 ^[2].

Although the increase in spinal hematoma in this patient population is not yet disastrous, it threatens to scuttle effective perioperative pain management. What series of critical incidents led to this situation? Did anesthesiologists underestimate the risk? Before the introduction of LMWH, many anesthesiologists regarded spinal hematoma as a rare, almost theoretical consideration. A review of the literature in 1994 by Vandermeulen et al. ^[3] identified only 61 cases of spinal hematoma associated with spinal or epidural anesthesia between 1906 and 1994. In addition, the safety of anticoagulation with IV (standard) heparin in association with spinal and epidural blockade has been well documented. Multiple studies had reported a zero incidence of spinal hematoma even with cardiopulmonary bypass and complete heparinization ^[4-6]. As a result of these investigations, guidelines for the regional anesthetic management of patients perioperatively anticoagulated with heparin were defined. For example, it is generally recommended that the needle and catheter be placed a minimum of 1 h before heparinization, the level of anticoagulation should be carefully monitored, and catheters should be removed when heparin activity is minimal ^[3,4,7]. Likewise, most anesthesiologists are adept at management of the patient with an indwelling epidural catheter while anticoagulated with warfarin. Although there is significant patient variability in the response to oral anticoagulants, the gradual onset of their effect allows the anesthesiologist to remove the catheter before therapeutic anticoagulation. However, as with heparin, the anticoagulant activity must be carefully monitored ^[8,9]. Reassured by the extensive safety record with standard heparin and warfarin, anesthesiologists concluded that neuraxial blockade in combination with LMWH would present no significant risk.

The allure of the anticoagulants themselves contributed to the maelstrom. LMWHs are highly effective agents of thromboprophylaxis. The drugs are administered subcutaneously; neither laboratory monitoring of the anticoagulant response to LMWH (anti-Xa levels) nor dose adjustment is required ^[10]. The simplicity of this regimen compared with oral anticoagulation with warfarin is obvious. However, these same properties complicate regional anesthetic management in the patient receiving LMWH. The lack of a conventional clotting assay to evaluate the level of anticoagulation during LMWH thromboprophylaxis forces the clinician to rely exclusively on pharmacology when estimating residual anti-Xa activity. In addition, because the LMWH dose is not adjusted for weight (with the exception of ardeparin), a relative overdose could occur in smaller patients. Finally, the pharmacologic differences between standard heparin and LMWH were not appreciated. The residual anticoagulant effect of LMWH is only partially reversed by protamine. The plasma half-life of LMWH is 2-4 times that of standard heparin and increases in patients with renal failure. Peak anti-Xa activity occurs 3-4 h after a subcutaneous LMWH injection, and 12-h anti-Xa levels are approximately 50% of peak levels ^[10]. Therefore, with twice daily dosing, there is significant anticoagulant activity even at trough.

A review of the experience in Europe, where LMWHs have been in clinical use since 1987, inspired false security. The first case of spinal hematoma in a patient receiving LMWH while undergoing regional anesthesia was reported in 1989 ^[11], with two additional cases in which the patients were also administered IV heparin and dextran noted in 1993 ^[2]. A review article by Bergqvist et al. ^[12] in 1992 estimated that more than one million patients had safely received LMWH in association with regional anesthesia and concluded that the combination seemed safe. No additional cases were reported in the European literature for several years; the risk of spinal hematoma in patients receiving LMWH was not considered clinically significant. However, a careful analysis of European literature between the years of 1993 and 1995 reveals that consistent practice guidelines among the European societies had been established ^[2,3,13]. Recommendations included a delay in needle placement of 8-12 h after a LMWH injection ^[2,3]. Subsequent administration of LMWH was postponed for 1-2 h ^[3] or, more conservatively, 8-12 h after needle placement ^[2]. Traumatic needle placement might result in an additional delay in LMWH administration or an alternate method of thromboprophylaxis. Likewise, the catheter was removed 8-12 h after LMWH administration ^[2,3] or 1-2 h before the next dose ^[3]. Formal guidelines for monitoring the patient's neurologic function were also developed.

Apparently, these recommendations have been successful. Only 11 spinal hematomas associated with LMWH and neuraxial block have been reported in the European literature since 1987. Although there may be unreported cases, the lack of discussion in European journals suggests that the anesthetic management strategies accurately assessed the problem and addressed it successfully ^[14]. It is important to realize that these recommendations were based on placing (and removing) needles and catheters during periods of decreased LMWH activity. This is fairly easily accomplished in Europe, where LMWH is administered on a once daily schedule and with a smaller total daily dose (40 mg) than that approved in the United States (30 mg twice daily).

The rarity of spinal hematoma makes identification of risk factors difficult. Even with 30 cases, only a partial evaluation is possible. However, examination of the patient and anesthetic variables suggests several possible risk factors. Approximately 75% of the patients were elderly women. An epidural catheter was placed in 22 patients; the first dose of LMWH was administered while the catheter was indwelling in 17 patients. Twelve patients received antiplatelet and/or warfarin therapy in addition to LMWH. Few patients developed neurologic symptoms while the catheters were indwelling. Although paralysis occurred shortly after catheter removal in at least seven patients, 24 h or more often elapsed between catheter removal and the onset of neurologic dysfunction (median time between initiation of LMWH therapy and neurologic dysfunction was 3 days). Importantly, the initial complaint was not severe radicular back pain, but new onset weakness or numbness. This is similar to findings in Vandermeulen et al.'s ^[3] study, in which muscle weakness and sensory deficit were reported as the first neurologic symptoms of spinal hematoma by 46% and 14% of patients, respectively. This atypical presentation may have delayed the diagnosis of spinal hematoma.

Anesthesiologists in the United States can draw on the European experience to develop their own practice guidelines for the management of patients undergoing spinal and epidural blockade while receiving perioperative LMWH. Although it is impossible to devise recommendations that completely eliminate the risk of spinal hematoma, preliminary analysis has identified possible contributory factors, which are addressed in the following recommendations.

1. The smallest effective dose of LMWH should be administered perioperatively. The FDA has recently approved enoxaparin 40 mg once daily as an alternate dosage regimen for thromboprophylaxis after total hip arthroplasty. Single daily dosing results in a true trough in anticoagulant activity, during which time needle placement and catheter removal can occur.

2. LMWH therapy should be delayed as long as possible: a minimum of 12 h and, ideally, 24 h postoperatively. This is within the FDA-approved dosage schedule and is particularly true for patients with indwelling catheters.

3. Antiplatelet or oral anticoagulant medications administered in combination with LMWH may increase the risk of spinal hematoma. Education of the entire patient care team is necessary to avoid potentiation of the anticoagulant effects.

4. The risk of spinal hematoma in patients with indwelling catheters who receive LMWH is almost certainly increased. Removal of the catheter before initiating LMWH therapy may be a compromise that provides superior analgesia for 24 h but eliminates the risk of concomitant epidural analgesia and LMWH therapy.

5. Catheter removal should occur when anticoagulant activity is low. Twice daily dosing of LMWH results in two peaks and no trough of anti-Xa activity. Skipping the evening dose of LMWH before an anticipated morning removal of the epidural catheter is unlikely to significantly increase the risk of thromboembolic events, but it should provide safe conditions for catheter removal.

Finally, all patients undergoing neuraxial block require repeated neurologic evaluations. A dilute local anesthetic or opioid solution permits the accurate assessment of neurologic status. Spinal hematoma should be suspected in any patient with a progression of neurologic deficits, such as motor weakness, numbness, bowel or bladder dysfunction, or new onset back pain. Neurosurgical intervention must be sought immediately, because recovery is unlikely if surgery is postponed more than 8 h ^[3].

The sinking of the Titanic resulted from a series of critical incidents, undetected or ignored, that escalated to a tragic conclusion. Recognition of the increased risk associated with spinal and epidural blockade and LMWH, continued surveillance and evaluation of the current information, and education are all crucial to averting future cases of spinal hematoma. Ultimately, this will require the combined efforts of the FDA, health-care providers, and industry, with leadership from professional societies. We must take the helm now or, as captains of neuraxial anesthesia and analgesia, we will surely go down with the ship!

Addendum

Additional cases of spinal hematoma associated with low molecular weight heparin and neuraxial block have been reported since this FDA Public Health Advisory was issued. Although the data are space for several of the events, as of April 15, 1998, nearly 40 cases of spinal hematoma have been recorded.