Preoperative Valproate Administration Does Not Increase Blood Loss During Temporal Lobectomy

Epilepsia, 37(1):9%101, 1996 Lippincott-Raven Publishers, Philadelphia 0 International League Against Epilepsy

Preoperative Valproate Administration Does Not Increase Blood Loss During Temporal Lobectomy Mariann M. Ward, Nicholas M. Barbaro, *Kenneth D. Laxer, and ?Ira J. Rampil Department of Neurosurgery, *Neurology, and fAnesthesia, University of California, San Francisco, California, U . S . A .

Summary: Surgical treatment is increasingly used for patients with medically refractory seizures. Valproate (VPA) is an effective, widely used anticonvulsant in this patient population, but believed by some researchers to increase surgical bleeding because of quantitative thrombocytopenia and functional defects in platelet aggregation. Because we have observed no clinical evidence that perioperative administration of VPA increases blood loss or complications related to postoperative bleeding in patients undergoing temporal lobectomy at our institution, we sought to test this hypothesis. We made a retrospective review of the medical records of all patients who underwent epilepsy surgery at the University of California, San Francisco Medical Center, from September 1986 through January 1993. Patients who had a temporal lobectomy and whose medical records documented preoperative platelet counts and pre- and postoperative hematocrit and hemoglobin values were included. We excluded pa-

tients who had cranial surgery before temporal lobectomy and those with intracranial neoplasms or vascular malformations. Patients were divided into two groups: those who received VPA in the immediate preoperative period and those who had not received VPA recently. We compared the estimated surgical blood loss and the estimated change in red blood cell (RBC) volume between groups by unpaired t tests. The charts of 87 consecutive patients qualified for inclusion in the study. Patients in the VPA group had relative (but not absolute) thrombocytopenia preoperatively (235 64 vs. 277 2 69 k in the No-VPA group). There were no differences in the estimated blood loss, RBC volume, or in the incidence of postoperative transfusion. VPA apparently does not increase complications of hemostasis during therapeutic surgical resections for epilepsy. Therefore, we do not recommend routinely discontinuing VPA before craniotomy . Key Words: Coagulopathy-Platelets-Temporal lobectomy-Valproate.

Surgical treatment has become the preferred therapy for an increasing number of patients with medically refractory seizures. Valproate (VPA) is an effective, widely used anticonvulsant in this patient population, but it may increase surgical bleeding due to quantitative thrombocytopenia (1-3) and functional defects in platelet aggregation (4). VPA may also decrease plasma coagulation factors (3, especially in pediatric patients (6), although this issue is controversial (4). Several case reports describe postoperative hemorrhagic complications in patients who received VPA (7,s). Consequently, some epilepsy centers discontinue VPA preoperatively . Other investigators discount the clinical significance of VPA-induced abnormalities in coagulation factors (5,9). Because we have observed no

complications indicating that perioperative administration of VPA increases blood loss or postoperative bleeding in patients undergoing temporal lobectomy at our own institution, we reviewed our experience with VPA in epilepsy surgery patients.

*

METHODS With the approval of our institutional review board, we reviewed the medical records of all patients who underwent epilepsy surgery at the University of California, San Francisco (UCSF) Medical Center from September 1986 through January 1993. Patients who had had a temporal lobectomy and whose medical records documented preoperative platelet count and pre- and postoperative hematocrit and hemoglobin values were included. Preoperative laboratory tests were obtained within 48 h of surgery. We excluded patients if they had had cranial surgery before temporal lobectomy or if they had intracranial neoplasms or vascular malformations. No patients were excluded for defects of

Received May 5, 1995; revision accepted August 1, 1995. Address correspondence and reprint requests to M. M. Ward at the Department of Neurosurgery, University of California, 505 Parnassus Avenue, San Francisco, CA, 94143-0112, U.S.A. Presented in part at the 1993 Annual Meeting of the American Epilepsy Society.

98

99

VPA AND SURGICAL BLEEDING

hemostasis. In keeping with our goal of tailoring temporal resection using electrocorticography (ECoG) to locate interictal spike activity, we routinely tapered the dose of all antiepileptic drugs (AEDs), including VPA, to approximately half of the patient’s baseline dose on the day before operation. VPA serum levels were not routinely obtained. All operations were performed by the same neurosurgeon. Intraoperative management was routine, and the decision to transfuse was guided by the UCSF Transfusion Guidelines, which recommend transfusion for hematocrit level S25. Patient records were divided into two groups for analysis: those from patients receiving VPA in the immediate preoperative period (VPA group) and those from patients who had not received VPA for >4 weeks (no-VPA group). Most of the no-VPA group had not received VPA for at least 3 months. Blood loss was determined in two ways: “estimated” and “calculated.” Estimated blood loss (EBL) was the quantity observed in the surgical field, in the suction waste bottles, and on the sponges, as recorded in the operative report at the conclusion of surgery. Calculated blood loss was based on changes in estimated red blood cell volume (VRBC) determined according to the method of Toy et al. (9); i.e., we assumed that total body blood volume (V,) was unchanged by surgery and thus VB values were the same preoperatively and 3 7 2 h postoperatively. Based on observations of Chaplin et al. (lo), Toy et al. further assume that venous blood is concentrated relative to the rest of the circulation and that a constant factor of 0.91 provides good correction to the average hematocrit level (Hct). Finally, we assumed that a single unit of blood contained 200 ml packed RBC. Pathophysiological conditions such as nephritis, cirrhosis, pregnancy, and splenomegaly, which might adversely affect the VRBCcalculation by altering the relationship between hematocrit and RBC volume, did not occur in our patient population. We defined RBC volumes according to the following formulas: VB

=

patient weight (kg) . 67.8 ml/kg for males and 62.3 ml/kg for females. Preoperative RBC volume = VRBCpreop = HCtpreop . vB Postoperative RBC volume = VRBCpostop = HCtpostop

’

B ‘

O*gl-

’

*

Oe91.

Then the percentage change in RBC volume is

The medical records provided documentation of the number of units of blood transfused and the occurrence of significant postoperative epidural or subdural hematoma. EBL and percentage loss of VRBc for the VPA and no-VPA groups were compared by unpaired t tests. The difference in the incidence of transfusion between groups was calculated by Fisher’s exact test. A value of p 6 0.05 was considered significant. We also used power analysis retrospectively to assess the likelihood of type I errors and, because of the observed population variances, to provide an estimate of the resolution of the comparisons. For power analysis, we judged a clinically meaningful difference in either EBL or AVRBC to be a 50% increase as compared with that observed in the control (no-VPA) group. RESULTS The charts of 87 consecutive patients qualified for inclusion in the study. The demographics of this population are shown in Table 1 . When stratified by the use of VPA, the groups showed no differences in age or gender composition. The minimum patient age was 14 years. The distribution of additional AED use is also shown in Table 1. For patients receiving VPA on admission, the mean dose was 31 k 1 1 mg/kg/day (range 6.25-1 17.3 mg/kg/day). The mean dose on the day before operation was 54 24% of the individual’s usual dose. VPA was completely discontinued on the day be-

*

TABLE 1. Study population characteristics (mean SD)

*

Parameter

No. VPA

VPA

Age (yr) Gender (M/F) Other AEDs Carbamazepine Phenytoin Clorazepate Primidone Phenobarbital Felbamate Mephenytoin Mephobarbital Lamotrigine Methsuximide Preoperaive hematocrit level Preoperative platelet count

30.7 f 9.1 23/28

28.5 t 9.5 16/20

32 21 2 7 4 1 1 1 I I 40.3 ? 4.0 277 2 69

21 9 6 1

2 1 1 0 0 0 40.3 2 4.6 235 t 64“

VPA, valproate; AEDs, antiepileptic drugs. “ p = 0.0078. Epilepsia, Vol. 37, N o . 1 , 1996

M . M . WARD ET AL.

100

fore operation in 9% of patients in the VPA group to maximize interictal spike activity during ECoG. Preoperatively, the VPA group on the average had decreased platelet counts (235 vs. 277 k in the VPA vs no-VPA group, respectively; p < O.Ol), and 1 patient in each group had moderate thrombocytopenia. The estimated and the calculated surgical blood losses did not differ between the VPA and no-VPA groups (Table 2), nor did the incidence of postoperative blood transfusion (Table 3). No patient received an intraoperative transfusion. Patients requiring postoperative transfusion had preoperative hematocrit levels and platelet counts that were not different from those of the nontransfused patients irrespective of VPA therapy. The transfused patients did, however, have greater blood loss than those who did not [EBL 450 vs 322 ml (p < 0.018); %AVRBC:986 vs. 327 ml (p < 0.000l)l. There was one instance of postoperative epidural hematoma in the VPA group, which was discovered 9 weeks postoperatively and drained with a small twist drill. The occurrence of a single epidural hematoma in 87 cases (1.14%) is similar to the reported incidence of 1 .O% after intracranial surgery (1 1). Using the observed blood loss variances in the study groups and our a priori estimate of meaningful change in blood loss, we calculated that the t tests we used had a statistical power of essentially 100% (p = in ruling out a type I error (12). Accepting a 90% power to exclude a type I error, the series of collected data would be adequate to detect a 63-ml difference in blood loss. With our current sample size, the power of comparison between rates of transfusion is -73%. Our study population was too small to determine whether the one instance of postoperative epidural hematoma in the VPA group constituted an actual difference between the two groups. DISCUSSION

Our study confirms the association of VPA therapy with mild preoperative thrombocytopenia. It also confirms our clinical impression that VPA therapy is not associated with an increased risk of complications of hemostasis during surgical resection TABLE 2. Estimated blood loss Observed EBL Calculated RBC loss Other sequelae

N o VPA

VPA

315 2 120 344 2 191 None

352 2 136 413 2 298 One epidural hematoma

VPA, valproate; EBL, estimated blood loss; RBC, red blood cell. Epilepsia, Vol. 37, No. I , 1996

TABLE 3. Patients requiring transfusion Transfusion No Yes

No VPA

VPA

49 2

32 4

VPA, valproate. p = 0.22 (two tailed) by Fisher’s exact test.

for epilepsy. A moderate decrease in preoperative platelet counts apparently does not increase surgical bleeding. Although reported data regarding the effect of mild to moderate preoperative thrombocytopenia are few, surgical bleeding in healthy young men with dilutional thrombocytopenia from massive transfusion does not increase until the platelet count decreases to