Ambulatory surgery is practiced worldwide. The shortage of nurses and the limitations of bed space, as well as the popularity of ambulatory surgery with patients, have encouraged the growth of ambulatory surgery and the emphasis on its efficacy . Since 60% of surgery in North America is done on an ambulatory basis , the study of adverse clinical outcomes after ambulatory anesthesia is essential.
Adverse outcomes, such as nausea, vomiting, dizziness, drowsiness, thirst, myalgia, sore throat, and delayed recovery have been reported after general anesthesia in ambulatory surgery patients [3-6]. Their incidence varies among studies. Fahy and Marshall  reported a 44.9% incidence of drowsiness, headache, and vomiting in outpatients. Dawson and Reed  found that the most common postoperative complications were nausea (30%), emesis (20%), and hypotension (10%). These adverse clinical outcomes may be associated with perioperative management, anesthetic technique, surgical procedure, duration of surgery, and fluid status of the patient. However, there is very little data about adverse clinical outcomes after the patient is discharged from the ambulatory center.
Since there is no prospective double-blinded randomized study of the effect of intravenous (IV) fluid therapy on adverse outcomes after ambulatory surgery, we studied the impact of perioperative IV fluid on adverse outcomes both at the hospital and at home after ambulatory anesthesia.
After obtaining institutional clinical ethics committee approval, and informed consent we studied 200 (ASA grade I-III) patients aged 18-55 yr, scheduled for ambulatory gynecologic, orthopedic, and general surgical operations. Exclusion criteria were history of valvular heart disease, previous congestive heart failure, preoperative nausea, vomiting or dizziness, intraoperative hypotension, and excessive blood loss. Demographic data, medical, social, and drug histories, and baseline hemodynamics were recorded prospectively. All patients fasted overnight; no sedative premedication was given.
Patients were randomly allocated to either a high-or low-infusion group. In each patient, an 18-gauge IV cannula was inserted under local anesthesia and 1% lidocaine. The high-infusion group of patients received IV Plasmalyte 148 isotonic (sodium chloride, 526 mg/100 mL; sodium gluconate, 502 mg/100 mL; sodium acetate, 368 mg/100 mL; potassium chloride, 37 mg/100 mL; magnesium chloride, 30 mg/100 mL) solution (20 mL/kg bolus) 30 min preoperatively in the ambulatory surgical unit; the low-infusion group received the same solution (2 mL/kg bolus) 30 min preoperatively. Once the desired volume of IV bolus was given, a new 1-L Plasmalyte 147 isotonic solution was started. Both the high- and low-infusion groups received 1 mL centered dot kg-1 centered dot h-1 during the intraoperative and postoperative periods. The IV cannula was removed 30 min after the operation in the post anesthetic care unit (PACU).
The investigator who assessed the symptoms and signs postoperatively, the attending anesthesiologist, and the PACU nurses were blinded to the patient-allocation group and to the amount of fluid infused preoperatively in the ambulatory surgical unit.
The total amount of fluid infused was documented. Patients treated with more than 1 mL centered dot kg-1 centered dot h-1 fluid intraoperatively and postoperatively because of hypotension or excessive blood loss were excluded from the study.
A standardized general anesthetic was administered. Orthopedic and laparoscopic patients were induced with 2-2.5 mg/kg of propofol and 1.5 micro gram/kg of IV fentanyl after precurarization with 3 mg of d-tubocurarine. Endotracheal intubation was accomplished with 1.5 mg/kg of succinylcholine. Anesthesia was maintained with a nitrous oxide (70%) and oxygen mixture (30%) with end-tidal isoflurane (0%-0.5%) in a semiclosed circle system. Vecuronium (0.1 mg/kg IV) was used for muscle relaxation and normocapnia was maintained. Neuromuscular block was reversed with glycopyrrolate (0.01 mg/kg) and neostigmine (0.05 mg/kg). Gynecologic patients were induced with 1 micro gram/kg of IV fentanyl and 2.5 mg/kg of IV propofol. Anesthesia was maintained with an infusion of 0.05-0.15 mL centered dot kg-1 centered dot min-1 IV propofol; patients were breathing spontaneously with a nitrous oxide (70%) and oxygen (30%) mixture. Electrocardiogram, blood pressure, heart rate, oxygen saturation, temperature, end-tidal CO2, airway pressure, tidal volume, minute volume, and end-tidal concentration of the inhaled anesthetic were monitored.
Heart rate, blood pressure in standing and supine position, and symptoms of nausea, vomiting, thirst, dizziness, and drowsiness were recorded by a blinded investigator at 30 and 60 min after surgery, and at discharge. Strict definitions were followed for the assessment of these symptoms. Nausea was defined as a volunteered complaint of nausea; vomiting, as active retching requiring antiemetic; dizziness, as faintness; vertigo, as gait disturbance 30 min after surgery; drowsiness, as sleepiness; and thirst, as a desire to drink. The time to first urination after arrival to PACU was noted. The time, dose, and route of all postoperative medications administered were noted. Home readiness was assessed with the postanesthesia discharge scoring system (PADSS; 7). This scoring system has five groups of physiologic variables: vital signs, activity and mental status, pain or nausea or vomiting, surgical bleeding, and intake and output. Each variable is scored from 0 to 2 Table 2. Patients were discharged when a score >or=to9 was reached. The time to achieve PADSS >or=to9 was recorded. Patients were interviewed by telephone on the first postoperative day: a standardized questionnaire was used Table 3.
The data was stored in dBase IV (Logical Training Systems, Inc., Rochester, NY) and was analyzed for statistical significance with Student's t-test and chi squared analysis, as appropriate. Results were expressed as means +/- SD; P < 0.05 was considered statistically significant.
Two hundred patients (15 males, 185 females) were randomly allocated to either the high-infusion (n = 100) or the low-infusion group (n = 100). The patients had a total of 15 orthopedic, 6 laparoscopic, 7 general surgical, and 172 dilation and curettage procedures. There was no significant difference between the high-infusion and low-infusion groups in age, sex, weight, and ASA classification, or type of surgery Table 1. No patient in the study received any extra fluid intraoperatively or in the PACU. Ninety-two percent of patients were successfully contacted for the 24-h post-operative telephone interview.
Total fluid infused was 1215 +/- 30 mL in the high-infusion group and 164 +/- 28 mL in the low-infusion group. There were no statistically significant differences between the two groups in the duration of anesthesia, the amount of anesthetic given, the number of patients who required postoperative pain medication, the amount of morphine given, or the postoperative requirement for antiemetic medication Table 1.
The incidence of thirst was significantly lower in the high-infusion group at all times than in the low-infusion group (P < 0.05; Figure 1). The incidence of drowsiness was significantly lower at all intervals in the high-infusion group Figure 2. Dizziness was significantly lower at all times except at discharge in the high-infusion group Figure 3. Nausea was significantly lower on the first postoperative day in the high-infusion group (P < 0.05; Figure 4). We did not find any significant difference in nausea at 30 min, 60 min, or at discharge between the two groups. There was no significant difference in vomiting between the two groups at 30 min, 60 min, at discharge, and 24 h postoperatively (not shown).
There was no significant difference between the groups in the time to first urination after arrival to the PACU (1.2 +/- 0.6 h vs 1.4 +/- 0.5 h) or the time taken to achieve PADSS >or=to9 (1.3 +/- 0.8 vs 1.5 +/- 0.4 h), or in the hemodynamic variables during the intra- or postoperative periods. Fifteen percent of patients in the high-infusion group complained of pain along the arm during infusion. There were no such complaints in the low-infusion group.
The incidence of adverse outcomes, such as nausea, vomiting, dizziness, drowsiness, thirst, and speed of recovery after surgery, depends on the surgical procedure, anesthetic technique and fluid status of the patient. Very little work has been done to determine the correlation between perioperative fluid therapy and the well being of the patients in the postoperative period. Cook et al.  found a trend toward improvement in these symptoms with fluid administration, especially with added sugar, but their study was not double-blinded. Our prospective double-blinded randomized study demonstrated that the incidence of adverse outcomes, such as thirst, dizziness, and drowsiness, was significantly lower in the high-infusion than in the low-infusion group at 30 min and 60 min after surgery, at discharge, and on the first postoperative day.
Adverse outcomes such as nausea, vomiting, thirst, drowsiness, and dizziness can create great distress in ambulatory patients. Nausea delays oral intake and worsens the general well being of patients. Retching because of nausea may increase pain and cause discomfort after minor abdominal surgery, such as laparoscopic procedures. Dizziness can precipitate nausea, vomiting, and restlessness and can delay ambulation. Postoperative drowsiness is potentially dangerous to patients if they cannot protect their airways. It also delays recovery and discharge.
These adverse outcomes delay early discharge and home readiness, thus increasing the workload of the nursing staff. Our findings indicate that these immediate adverse outcomes after ambulatory surgical procedures persisted for more than 24 h, but there was no significant difference between the two groups in the time to achieve PADSS. Hydrating the ambulatory surgical patient is therefore advantageous because it reduces the incidence of adverse clinical outcomes. Perioperative dehydration leaves a considerable deficit of fluid in surgical patients. Overnight fasting in a 70-kg patient undergoing surgery the next morning is likely to cause at least a 1-L fluid deficit . Normal insensible perspiration in an adult with a normal temperature amounts to a per-hour fluid loss of approximately 0.5 mL/kg body weight; which is approximately equal to the fluid lost by the body through urine. The study methodology was double-blinded. Neither the attending anesthesiologist nor the nursing staff was aware of the study group allocations. Therefore the extra IV fluid in the high-infusion group was given preoperatively. During a surgical procedure, however, the water lost through perspiration can increase considerably because of unhumidified anesthetic gases. Keane and Murray  found a statistically significant reduction in serum osmolality in a group of patients who received perioperative fluids . Furthermore, Shires et al.  reported that the functional extracellular fluid volume was reduced during minor as well as major surgical procedures. An increase in antidiuretic hormone secretion in response to dehydration and stimulation of osmoreceptors also partly contributes to a postoperative reduction in urine output .
Different intraoperative regimens of fluid administration have been proposed, depending on the type of surgery and third-space fluid loss. Jenkins et al.  suggested 12-15 mL/kg for the first hour and 6-10 mL/kg for the next 2 h. Fluid is administered to keep urine output at 1 mL centered dot kg-1 centered dot h-1, or 50-100 mL centered dot kg-1 centered dot h-1 for a major operation from the third hour onward. Davidson  indicated that if the preexisting fluid and electrolyte abnormalities have been corrected before the operation, fluid administration should maintain the urine volume between 0.5 and 1 mL centered dot kg-1 centered dot h-1. Campbell et al.  observed that renal and cardiovascular stability is much better when crystalloids are given at the rate of 10-15 mL centered dot kg (-1) centered dot h-1 intraoperatively.
There has been no standardized fluid regimen available for clinical use in ambulatory patients. Patients selected for the study were usually studied during the morning session, and at worst would endure a fast of approximately 16 h, from 9 PM the previous night until 1 PM the afternoon after surgery. The fluid load of 20 mL/kg was based on a daily water requirement of approximately 30 mL/kg. We found that 20 mL/kg of isotonic electrolyte solution reduced the adverse clinical outcomes of thirst, nausea, dizziness, and drowsiness. Our patients had nothing to drink after midnight. Scarr et al.  have recommended that healthy patients be allowed to drink clear fluids until 3 h before their elective ambulatory surgery is scheduled. Further study to determine the optimal perioperative fluid replacement in ambulatory patients who drink up to 3 h preoperatively is warranted.
In conclusion, our results showed that alleviating dehydration with adequate fluid therapy reduced the incidence of postoperative adverse outcomes, such as thirst, nausea, dizziness, and drowsiness. We recommend perioperative hydration of 20 mL/kg for fluidrestricted patients undergoing general anesthesia in short ambulatory surgery.
The authors thank Dr. Keith Rose for his constructive suggestions with the study design.
1. Kitz DS, Slusary-Ladden C, Lecky JH. Hospital resources used for inpatients and ambulatory surgery. Anesthesiology 1988;69:383-6.
2. Orkin FK. Economic and regulatory issues. In: White PE, ed. Outpatient anesthesia. New York: Churchill Livingston, 1990:87-106.
3. Fahy A, Marshall M. Post anesthetic morbidity in outpatients. Br J Anaesth 1969;41:433-7.
4. Dawson B, Reed WA. Anaesthesia for day-care surgery: anaesthesia for adult surgical outpatients. Can Anaesth Soc J 1980;27:409-11.
5. Natof HE. Complications associated with ambulatory surgery. JAMA 1980;244:1116-8.
6. Brindle GF, Soliman MG. Anaesthetic complications in surgical outpatients. Can Anaesth Soc J 1975;22:613-9.
7. Chung F. Discharge criteria, is it changing? J Clin Anesth 1993;5:64S-68S.
8. Cook R, Anderson S, Riseborough M, Blogg CE. Intravenous fluid load and recovery. Anaesthesia 1990;45:826-30.
9. Keane PW, Murray PF. Intravenous fluids in minor surgery. Their effect on recovery from anaesthesia. Anaesthesia 1986;41:635-7.
10. Thoren L, Wiklund L. Intraoperative fluid therapy. World J Surg 1983;7:581-9.
11. Shires T, Williams J, Brown F. Acute change in extra cellular fluids associated with major surgical procedures. Ann Surg 1961;154:803--10.
12. Moran WH, Zimmerman B. Mechanisms of ADH control of importance to the surgical patient. Surgery 1967;62:639-44.
13. Jenkins MT, Giesicke AH, Johnson ER. The postoperative patient and his fluid and electrolyte requirements. Br J Anaesth 1975;47:143-50.
14. Davidson GM. Intraoperative fluid and electrolyte requirements. Anaesth Intensive Care 1977;5:333-8.
15. Campbell IT, Bater JN, Tweedie IE, et al. IV fluids during surgery. Br J Anaesth 1990;65:726-9.
© 1995 International Anesthesia Research Society
16. Scarr M, Maltby JR, Jani K, Sutherland LR. Volume and acidity of residual gastric fluid after oral fluid ingestion before elective ambulatory surgery. Can Med Assoc J 1989;141:1151-4.