html xmlns=”http://www.w3.org/1999/xhtml” xmlns:mml=”http://www.w3.org/1998/Math/MathML” xmlns:epub=”http://www.idpf.org/2007/ops”>
Chapter 19 Spinal anesthesia
Fluid therapy is widely used in conjunction with spinal anesthesia to minimize hypotensive events. The use of crystalloids for this purpose seems to be only minimally effective, particularly when given prior to the administration of the spinal anesthetic. To be effective, substantial fluid boluses must be administered – of the order of 20 ml/kg – and then preferably as a rapid coload simultaneously with the induction of spinal anesthesia. Several studies and meta-analyses suggest that colloids, either as preload or coload, are more effective than crystalloids and may result in a smaller volume of fluid loading being required. However, fluids alone, whether crystalloid or colloid, are generally inadequate to prevent or treat significant hypotension associated with spinal anesthesia, and the concomitant use of a vasopressor will frequently be necessary, particularly in obstetrics. The best that can be achieved with optimal fluid therapy is an overall reduction in the total dose of vasopressor required. The best available management of spinal hypotension would appear to be optimal fluid therapy combined with carefully graded administration of the appropriate vasopressor. It is possible that goal-directed fluid therapy, using an appropriate analysis of cardiac performance to assess the response dynamic indices to a fluid challenge, may improve fluid therapy in the future, but, at present, the evidence for this is insufficient to make a firm recommendation.
Introduction
Spinal anesthesia is inevitably associated with some degree of hypotension. Estimates of the incidence of clinically relevant hypotension vary from 25% to over 80%, depending on the criteria used to identify clinically relevant decreases in blood pressure, and on the population group studied.
Two groups of patients have been extensively studied. The first patient population of concern comprises elderly patients undergoing a variety of surgical procedures for which spinal anesthesia is appropriate, including bladder and prostate procedures and lower limb surgery, particularly joint replacements. The second important group is the obstetric population, in whom hypotension has implications for the fetus (reduction in placental perfusion) and for the mother (cardiovascular collapse and unpleasant side effects of nausea and vomiting).
A variety of causative factors have been suggested to explain the hypotension following spinal anesthesia. These include diminished cardiac output as a result of inadequate compensation for arteriolar- and venodilatation in elderly patients with impaired ventricular function, and an only partial compensatory increase in cardiac output in obstetrics patients with normal ventricular function. There may also be paralysis of the sympathetic nerve supply to the heart and adrenal glands resulting in reduced catecholamine responsiveness and unmasking of hypovolemia. In the pregnant patient these may be complicated by aorto-caval compression. Pre-existing hypertension may increase the risk of spinal anesthesia-associated hypotension in older patients,[1] but pre-eclampsia is associated with less spinal hypotension than in healthy parturients, since the usual cardiovascular status is one of increased inotropy together with elevated systemic vascular resistance.[2]
Older texts refer to the value of electrolyte solutions in minimizing the hypotensive response and suggest that volumes of 2 liters may be adequate. These assumptions have been challenged by a substantial body of research over the past 20 years.
Non-obstetric spinal anesthesia
In a study conducted in elderly patients aged 60 years or over, the overall incidence of arterial hypotension during spinal anesthesia was 27%, rising to 60% when temperature sensation was blocked to T7. If the block reached T4, all patients required vasopressor therapy. Patients were divided into three groups receiving 16 ml/kg, 8 ml/kg, or zero acetated Ringer’s lactate solution (RL) as a preload; crystalloid preloading had no effect on the incidence of hypotension.[3] However, a relatively small preload (7 ml/kg) of 3% hypertonic saline was significantly better than 0.9% saline at reducing the requirement for vasopressor support in patients undergoing prostatectomy.[4] RL was shown to sustain, but not increase cardiac output in healthy patients undergoing lower extremity surgery.[5]
Fluid kinetics modeling has shown that crystalloids rapidly redistribute away from the central compartment, and tend to enhance the second compartment. The redistribution time constants of crystalloid are quite rapid,[6] and these kinetics may contribute to the relative lack of efficacy of crystalloid solutions in preventing spinal anesthesia-associated hypotension. Volunteer studies have shown that an infusion rate of 50 ml/min is required to yield an increase in blood volume of approximately 10%, and that this would need be maintained for at least 40 min to produce the required volume expansion; this volume load rapidly dissipates.[7] In a review of spinal anesthesia in elderly patients, Critchley concluded that an adequate venous preload was necessary, but any fluid loading should be ideally administered as the block is developing (subsequently termed coload in an obstetric anesthesia study [8]), rather than as a preload.[9] In an early study of volume kinetics during epidural anesthesia, it was shown that a crystalloid coload of 15 ml/kg was better retained within the circulation in those patients who showed a substantial decrease in systolic pressure. Nevertheless, despite volume loading, there was a relative hypovolemia, as evidenced by hemodilution, in all patients throughout the development of hypotension.[10] A recent study using either crystalloid (1,000 ml) or colloid (500 ml) coload in older patients showed that both forms of fluid therapy sustained cardiac output above baseline values for 30 min, but the effect of crystalloids waned after 20 min.[11]
Goal-directed fluid therapy
Levy and colleagues recommended the use of goal-directed fluid therapy with colloids together with spinal (but not epidural) anesthesia to optimize patient outcomes after laparoscopic surgery.[12] In a subsequent study, patients undergoing laparoscopic colo-rectal surgery were randomized to receive spinal or epidural analgesia or intravenous morphine. All patients were treated with volume optimization against esophageal Doppler measurements. Patients who failed to achieve an indexed oxygen delivery of >400 ml/min/m2 had a higher rate of anastomotic leak than those achieving this level, but the fluid therapy itself and the mode of analgesia were not individual predictors.[13] A recent study examined the effects of goal-directed fluid therapy using the LiDCO® device in patients undergoing hip surgery and found no evidence of improved outcome relative to a standard care group, and no evidence of benefit. These authors concluded that there was insufficient evidence either to support or discount routine use of goal-directed fluid administration in these patients.[14] In a clinical study of isobaric spinal anesthesia in patients undergoing transurethral bladder tumor resection, patients receiving crystalloid (15 ml/kg) showed a significantly longer mean time to reach the peak sensory block and a lower median sensory block at 15 and 20 min than those receiving colloid (5 ml/kg). In the crystalloid group, cerebrospinal fluid (CSF) flow in the cranial direction decreased significantly and attenuated pulsatile movement of CSF at the L2–3 intervertebral intrathecal space. However, this was not observed with colloid.[15]
Obstetric spinal anesthesia
The management of spinal hypotension in obstetrics is crucial for the safety and comfort of the mother, and the well-being of her baby.
An early obstetrics study showed that the administration of a 20 ml/kg crystalloid preload failed to prevent significant hypotension whether the load was given rapidly over 10 min or more slowly over 20 min. However, in the rapid preload group, three patients were found to have a marked rise in central venous pressure. The authors queried the role of isotonic crystalloid preload in the management of spinal anesthesia for elective Cesarean section,[16] and the same group recommended that if time was scarce the spinal anesthesia should not be delayed by awaiting administration of the crystalloid preload.[17] Jackson and colleagues evaluated the use of a preload of 1,000 ml prior to spinal anesthesia for Cesarean section and were unable to demonstrate any advantage over a preload of 200 ml. They also concluded that crystalloid preload was not of value.[18]
A study in pre-eclamptic patients has also failed to demonstrate significant benefit from a preload of 1,000 ml crystalloid, although the authors commented that changes in uterine artery velocity waveforms were minimal and there was no adverse effect on the neonate.[19] There have been a number of concerns regarding the possible adverse consequences of saline-based solutions given in large volumes in terms of generating hyperchloremic metabolic acidosis, but this does not seem to be of particular consequence to either mother or child when saline solution is compared with lactated Ringer’s for preload,[20] and this has since been confirmed in a study of starch in a balanced salt solution compared with a similar starch in saline.[21]
Crystalloid coload
The relative lack of efficacy of crystalloid loading prior to spinal anesthesia can be explained by a number of factors including the physiological responses of patients with normal fluid balance status to a rapid fluid load, and the volume kinetics of crystalloid solutions. Pouta and colleagues demonstrated a significant increase in the release of atrial natriuretic peptide, and a lesser effect on endothelin-1, following crystalloid loading of 2,000 ml lactated Ringer’s solution in healthy parturients. They concluded that this could offset the effects of volume load on blood pressure during Cesarean delivery.[22] This release of hormone was correlated significantly with the increase in atrial stretch as indicated by an increase in central venous pressure. Pre-eclamptic patients showed a greater response, possibly in line with reduced diastolic function in these patients.[23]
A dose-defining study of 90 parturients looked at the effect of three fluid volume groups receiving 10, 15, or 20 ml/kg of RL respectively within 15 min before the spinal block. Spinal anesthesia was followed immediately by an infusion of ephedrine of 3 mg/min in all groups. The incidence of hypotension was 60%, 36.7%, and 13.4% in groups 10RL, 15RL, and 20RL, respectively (p < 0.05). Additional ephedrine dosage was lowest in group 20RL compared with the other groups (p < 0.05). The incidence of nausea and vomiting in group 20RL was significantly less than in group 10RL (p = 0.02). It was concluded that preloading with 20 ml/kg of RL prior to spinal anesthesia followed by an ephedrine infusion reduced the incidence of hypotension and of nausea and vomiting, and decreased the total dose of ephedrine.[24]
Since the administration of crystalloids to healthy, volume-replete subjects is likely to trigger physiological responses such as secretion of atrial natriuretic peptide and increased renal excretion of the fluid, and volume kinetics suggest very rapid redistribution of fluid loads in the absence of a hypovolemic state, the administration of fluids as a coload at the time of onset of spinal anesthesia might be more efficacious than administering a preload. Initial studies using 20 ml/kg of crystalloid as a rapid bolus at the time of the performance of spinal anesthesia suggested that a significant reduction in ephedrine requirements could be achieved by this method.[8]
A meta-analysis of preload vs. coload failed to confirm the advantage of coload, although the authors commented that the only study in which coload was effective was the one that used the highest infusion volume, i.e. 20 ml/kg.[25] Fluid kinetic studies have shown that the distribution of crystalloids from the plasma to the interstitium is markedly delayed during the onset of spinal, epidural, and general anesthesia,[26] and this lends support to the concept that coload with crystalloids may be more effective than preload, but definitive data are lacking. There is also no evidence that colloid coload is more effective than colloid preload.[27]
Colloids
The final issue to be resolved is whether the use of colloid solutions would be advantageous by virtue of their retention in the circulating blood volume for a longer period than crystalloids.
Studies on the effects of spinal anesthesia on cardiac output suggested that volume preloading had to be sufficient to produce a significant increase in cardiac output if hypotension was to be minimized.[28] These authors showed that colloid preload was more effective than crystalloid in enhancing cardiac output and, consequently, in reducing spinal hypotension during elective Cesarean section.
Whilst logic would suggest that colloid preloading would be advantageous compared with crystalloid loading, the data on this topic are inconsistent. Riley and colleagues suggested that a preload of 500 ml hydroxyethyl starch (HES) together with 1 liter RL was superior to 2 liters of RL in terms of reductions in the incidence of hypotension and requirements for ephedrine in obstetrics patients.[29] However, a study found that 500 ml of gelatin colloid preload was not superior to a similar volume of crystalloid or no preload at all in parturients.[30] The same group also failed to show a benefit of a combination of 500 ml each of HES and crystalloid in elderly patients receiving spinal anesthesia.[31]
Obstetrics studies using 1 liter colloid showed significantly less hypotension than no preload,[32] or preloading with either 1.5 liters lactated Ringer’s or 500 ml HES.[28] A systematic review at this time concluded that crystalloid preload was inconsistent in preventing hypotension, whereas colloid was generally effective in minimizing hypotension, but neither was effective in minimizing maternal nausea, and there were few differences in neonatal outcomes.[33] Dahlgren and colleagues confirmed that 1 liter of colloid was more effective than an equivalent volume of crystalloid for the prevention of hypotension [34] and subsequently demonstrated that supine stress testing could accurately predict those patients in whom a colloid preload would likely be beneficial.[35] Davies and French demonstrated that 10 ml/kg of colloid was more effective than 5 ml/kg at minimizing obstetric spinal hypotension.[36]
A comparison between combinations of 1 liter each of RL combined with gelatin or HES-based colloids demonstrated that the HES-RL was superior to either the gelatin combination or 1 liter HES alone.[37] A recent Cochrane review concluded that crystalloids were more effective than no fluids (relative risk [RR] 0.78, 95% confidence interval [CI] 0.60–1.00; one trial, 140 women, sequential analysis) and colloids were more effective than crystalloids (RR 0.68, 95% CI 0.52–0.89; 11 trials, 698 women) in preventing hypotension following spinal anesthesia at Cesarean section.[38] Although no differences were detected for different doses, rates, or methods of administering colloids or crystalloids, the literature review presented above suggests that at least 1 liter of colloid is required to produce a significant reduction in the incidence of hypotension in healthy patients.
In terms of cardiac output, a study using Doppler monitoring during 1 liter coload with crystalloid or colloid found similar cardiac outputs between the groups, with no differences in secondary outcomes of vasopressor use or hemodynamic stability.[39] However, a subsequent meta-analysis involving 10 trials with 853 parturients (and including the previous study) concluded that, when colloid was used, there were significantly fewer hypotensive events (odds ration (OR) 3.21, 95% CI 2.15–4.53, number needed to treat = 4), less demand for vasopressors (standard mean difference SMD 0.77, 95% CI 0.34–1.21) and improved cardiac output (SMD −1.08, 95% CI −2.00 to −0.17).[40]
A study has recently been published comparing 500 ml of 6% HES (130/0.4) + 500 ml of RL (HES group) or 1,000 ml of RL (RL group) i.v. before spinal anesthesia. The incidence of both hypotension and symptomatic hypotension was significantly lower in the HES group than in the RL group. There was no detectable placental transfer of HES in six umbilical cord blood samples analyzed in the HES group, and neonatal outcomes were comparable.[41]
In reviewing the current trends, Mitra et al. concluded that intravenous crystalloid pre-hydration is not very efficient and that the focus has changed toward co-hydration and use of colloids.[42]
Related posts:
Full access? Get Clinical Tree
