S179 INTRODUCTION: Intraoperative hypovolemia is common and is associated with an increased postoperative morbidity and prolonged hospital stay [1]. Esophageal Doppler (EDM Deltex Med Inc., Irving, Texas) allows non-invasive continuous monitoring of stroke volume (SV) and calculated cardiac output (CO), thus enabling a continuous measure of volume status. We investigated whether optimizing plasma volume with the use of EDM would reduce postoperative morbidity and length of hospital stay. METHODS: Following IRB approval and informed consent, patients undergoing "moderate risk" surgical procedures (major urological, major gynecological with laparotomy, and major general surgical with laparotomy) with an anticipated blood loss >500 ml were studied. All patients received a standardized balanced anesthetic technique. After induction of anesthesia, an EDM probe was inserted via the patient's mouth and positioned in mid-esophagus. Patients were randomized into therapy (T) or control (C) groups. In the T group, intraoperative fluid management was guided by a protocol with 200 ml aliquot of Hespan[trade mark sign] (up to a total of 20 ml/kg) given to obtain a maximum stroke volume and corrected flow time. This procedure was repeated every 15 min until the end of surgery. Lactated Ringer's was used when the Hespan[trade mark sign] maximum volume (20 ml/kg) was reached. Patients in the control group have their fluid managed based on clinical criteria. The EDM parameters were not displayed in the C group (the monitor screen was covered with an opaque card). Data on intraoperative hemodynamics, hospital stay and morbidity were collected. Wilcoxon rank sum test and Fisher exact test were used to analyze data. RESULTS: Forty four patients (Therapy = 21, Control = 23) have completed the study to date. Patient demographics, type and duration of procedures were comparable between the groups. The intraoperative fluid administered, hemodynamic data and recovery data are shown in Table 1 below. One patient in the control group developed postoperative renal dysfunction. There was no difference in blood or blood product usage between groups.Table 1DISCUSSION: Intraoperative fluid optimization guided by EDM appears to be associated with faster recovery and shorten hospital stay.

General anesthesia during pregnancy has been associated with an increased incidence of spontaneous abortion, low birth weight, and premature delivery [1,2]. In addition, multiple pharmacologic drugs may be associated with fetal toxicity and teratology when administered during the perioperative period [3]. Although the most prudent management is to avoid elective surgery in the parturient, certain diseases require emergent or urgent surgery during pregnancy, and a complementary anesthetic technique must be used to minimize these potential risks. The management of breast carcinoma during pregnancy is complex, and it may necessitate surgery and anesthesia. In addition to the specific anesthetic implications of pregnancy, mastectomy is associated with an increased incidence of postoperative nausea and vomiting as well as difficult pain management [4]. Specific recommendations regarding this disease in parturients are not readily available. Paravertebral nerve blockade (PVB) has been used at our institution to achieve effective surgical anesthesia and postoperative pain control for breast cancer surgery [5]. In initial trials, this technique offered prolonged analgesia and a decreased incidence of nausea and vomiting. We report the use of PVB in a pregnant patient requiring modified radical mastectomy (MRM) with axillary dissection. A 38-yr-old primagravida at 29 wk gestation presented with a 10-cm left breast mass. A biopsy revealed an infiltrating ductal carcinoma grade 3 that required urgent left MRM. Two weeks later, she was scheduled for a MRM. Her pregnancy had been complicated by gestational diabetes. The rest of her medical history was unremarkable. Preoperative physical examination revealed a gravid female, heart rate 82 bpm, blood pressure 120/80 mm Hg, respiratory rate 20 breaths/min, and fetal heart rate 120-140 bpm with good beat to beat variability. The remainder of her physical examination was unremarkable. The patient was brought to the preoperative holding area, where she received 30 mL of oral sodium citrate. Routine noninvasive monitors, including a fetal heart rate monitor, were placed. IV access was obtained, and supplemental oxygen was administered. The patient received titrated doses of iv midazolam and fentanyl for moderate sedation. With the patient in the sitting position, PVB was performed at T1-6 as described by Greengrass et al. [5] and Greengrass and Steele [6]. Four milliliters of 0.5% ropivacaine was administered at each level. The patient was then placed in the supine position with left lateral uterine displacement. The patient tolerated the procedure well, and there were no fluctuations in maternal or fetal vital signs that required treatment. After PVB, sensory blockade was confirmed by loss of sensation to pinprick over the T1-6 dermatomes on the left anterior chest wall, compared with the intact sensory noted over the right chest wall. The patient was sedated using an iv propofol infusion of 25 [micro sign]g [middle dot] kg-1 [middle dot] min-1 for anxiolysis. The surgery proceeded uneventfully. The patient remained comfortable during the surgery and no anesthetic interventions were required. Postoperatively, the patient remained comfortable, with no complaints of nausea or vomiting. The patient only requested supplemental analgesics once the morning after surgery, approximately 18 h after neural blockade. No additional narcotics were required. The patient was discharged on Postoperative Day 2, pain-free and without complaints. The patient was readmitted 4 wk later with a fetus in frank breech presentation, underwent an uneventful cesarean section, and delivered a healthy, full-term newborn. The patient was immediately referred to her oncologist for completion of her multimodal cancer therapy. Discussion Surgery occurs in an estimated 2.0% of all pregnancies [1]. Shnider and Webster [7] reported a perinatal mortality rate of 7.5% in 147 women who underwent surgery during pregnancy. Duncan et al. [2] found that general anesthesia administered during the first and second trimester was associated with the highest incidence of spontaneous abortion after gynecologic procedures. The intense emotional anguish of a breast cancer diagnosis may require the need for anxiolytics to suppress increased maternal catecholamines. Increased endogenous catecholamines may contribute to decreased uterine blood flow and may potentiate fetal hypoxia. Midazolam was used in this case in the late third trimester because it is a short-acting, water-soluble imidazobenzodiazepine that can produce anterograde amnesia [8]. Although studies condemn the use of anxioloytics during early pregnancy, investigations of the placental transfer of midazolam when given as a premedicant before elective cesarean section support the use of midazolam in obstetrics [9,10]. There may be a dramatic sensitivity to exposure during certain days of gestation but little or no effect later in gestation. Davies et al. [11] estimated the incidence of failed intubation in parturients to be 1 in 500. Using regional anesthetic techniques is one option to avoid instrumenting the airway and to minimize exposure to anesthetics. In fact, the heightened awareness of local anesthetic toxicity and the use of regional anesthesia may contribute to lower the overall incidence of maternal mortality [12]. Traditionally, high thoracic epidural anesthesia has been the alternative to general anesthesia for surgery of the thorax, chest wall, and axilla. However, despite providing good operative conditions, thoracic epidural blockade can be associated with hypotension, bradycardia, and cardiac arrest from bilateral sympathetic, blockade and other physiologic reflexes [13,14]. In addition, postoperative analgesia via a thoracic epidural continuous infusion requires increased monitoring in the postoperative period. Analgesia from an epidural technique is also limited when the infusion is discontinued, which is not the case for PVB. PVB requires no continuous infusion to maintain postoperative analgesia for a mean of 18 h when bupivacaine 0.5% with epinephrine was the local anesthetic of choice for the initial series described by Coveney et al. [15]. Potential complications of PVB, as with many regional anesthetic techniques, include inadvertent needle puncture of adjacent structures. Lateral insertion can result in a pneumothorax, whereas medial placement can result in epidural or spinal anesthesia. As with other techniques involving local anesthetics, care should be taken to avoid accidental intravascular injection [6]. In our experience, symptomatic pneumothorax has complicated care in only one patient in >600 patients undergoing PVB. This patient was managed conservatively without the use of a chest tube. In a large multicenter study, Lonnqvist et al. [16] demonstrated a success rate of >90% with a similar technique and found complication rates similar to those of other commonly used regional anesthetics. The frequency of complications for PVB reported by Lonnqvist et al. [16] was hypotension 4.6%, vascular puncture 3.8%, and pneumothorax 0.5%. In this case, the patient had 18 h of analgesia PVB with no complaints of nausea or vomiting. This is encouraging because postoperative nausea and vomiting (PONV) after breast cancer surgery has been reported to affect as many as 59% of women receiving a general anesthetic.1 The management of postoperative pain after breast cancer surgery with narcotics contributes to a cycle promoting PONV, which may prolong the hospital stay [4,17]. These data are important because the management of postoperative pain and PONV during pregnancy may be difficult and may require multimodal regimens, which can affect the fetus. (1) Miguel R, Rothschiller J, Majchrzak J. Breast surgery is a high risk procedure for development of nausea and vomiting [abstract]. Anesthesiology 1993;79:1095. Breast cancer surgery during pregnancy presents considerations that are different from the conventional management of this disease. Physiologic changes of pregnancy, as well as the presence of the fetus, limit the scope of anesthetic techniques. An anesthetic technique that provides maintenance of airway reflexes, limits physiologic changes, blocks the nocioceptive response, and provides prolonged analgesia with minimal side effects is ideal. PVB provides an alternative method of management to achieve these goals, is simple, and may offer advantages over current choices.