Blog with interesting cases and/or problems related to anesthesia with discussion based on best evidence in the literature.

March 8, 2011

preload or not to preload

Recently a fellow anesthesiologist questioned my judgement when he discovered that I had not given a preload of crystalloid to my patient prior to intiating labor analgesia. Therefore, I present the historical basis for the common practice of preloading patients for labor epidurals when there is no evidence to support the practice. I did quiz the OB nurses regarding the practice and they stated that while there was no policy requiring this, most anesthesia providers give them a hard time about it, so most will do it without any order from the physician, and the labor unit has a culturally imposed bias towards preloading. This medical intervention then bypasses the MD's wishes in regards to the patient. Is this bad or good? In reality, it is probably neither in most cases. In some cases it may be good if the laboring patient happens to be dehydrated and thus at risk for severe hypotension. However, in some cases it may lead to fluid overload, decreased uterine contractions and unnecessary interventions, or worsen dependent edema.

Initial studies done looking at this issue were performed in gravid ewes. In this model it was shown that an IV fluid bolus was effective in reversing hypotension following spinal anesthesia. Thereafter, studies were performed in human subjects. The practice of preloading was probably given a strong foothold after Collins et al. found that a preload of 1 liter of hartmann's s0lution prior to initiation of labor epidurals decreased the incidence of hypotension from 28% to 2%. They also found that FHT deterioration was lessened by in the preloaded women. However, they used 0.375% bupivacaine in their loading dose, significantly more than presently used. This study was performed in 1978, and current practice rarely if ever finds an anesthesiologist giving 0.375% bupivacaine to initiate labor analegesia. More recent studies using contemporary methods have not found a benefit to preloading prior to labor analgesia. Zamora et al. [5] compared to fluid preload regimens (500mL LR vs 1L LR). They titrated lidocaine to a T10 level and found the incidence of hypotension to be 4% in both groups. They also compared FHT abnormalities. The incidence of was low and not different between groups. The only difference, was that in patients who received 1L of LR there was a significant decrease in uterine contractions and a delay in the placement of the labor epidural. In another study, bupivacaine 0.25% was used as an initiation dose, and the authors were not able to find a difference between preload vs. no preload. Two studies came out in 2000. Kinsella et al. tried to recreate in many ways the collins study by looking at the same number of women that collins et al. used in their study and looking at the same end points. Kinsella et al. used a lower concentration with fentanyl to initiate labor epidural analgesia and did not see a decrease in the incidence of hypotension in women who received a 1 L preload of crystalloid. FHTs were also recorded. While their was a trend toward FHT deterioration in the group that did not receive a preload, this was not statistically significant and they noted that they would need 200 more women in order to have sufficient power to determine this. That same year, an article was published in the European Journal of Anesthesiology looking at preloading and its benefits (or lack thereof) in patients who had received a CSE with local anesthetics and narcotic. The preload was only 500 mL LR. They did not find any differences between groups. The following year (2001), Hawthorne et al. published an article in the International Journal of Obstetric Anesthesia. They studied 30 women who either received 7ml/kg NS or no preload and received a labor epidural with 0.1% bupivacaine and 2mcg fentanyl at 20 mL to iniate. In this study there was no difference between groups. In 2003, in the same journal another group (Kubli M et al.) used a very similar protocol as that of Hawthorne in 2001. However, they compared just over 160 women, one group who received no preload and the other who received 7ml/kg of crystalloid. The incidence of hypotension was low in both grous (11% vs. 13%) and not different. FHT were also considered and abnormalities occurred in 20% of patients not given a preload vs. 15% of those given 7 ml/kg of crystalloid (not statistically different). Of course, in the above studies, both dilute local anesthetic solutions as well as small preloading doses were used. Rout el al. looked at spinal anesthesia for cesarean sections and used larger volumes of crystalloid (20ml/kg) preload. While hypotension was greater in the no preload group (71% vs. 55%), their study demonstrated that even with large volumes of crystalloid, hypotension could not be avoided in more than half of their patients. They also reported that the severity, timing and duration were not different, and more importantly, neonatal conditions were not different between groups. Another author, Ueyama et al. looked at the effects of crystalloid vs. colloid on blood volume and cardiac output to determine the relationship of these with the development of hypotension. They discovered that hypotension could only be avoided in women who received enough volume to cause a significant increase in blood volume and cardiac output. This was best accomplished in the group receiving 1 L of colloid vs 1L crystalloid or 0.5 L colloid (hypotension was 75%, 58% and 17% respectively). In 2001, Morgan et al. published a qualitative systematic review to see how crystalloid compares to colloid at preventing hypotension prior to spinal anesthesia for cesarean section. The review concluded that giving a crystalloid preload resulted in hypotension twice as often as those who had received a colloid preload.
In the end, the traditional policy of giving a 1 L preload of crystalloid is very unlikely to increase blood volume significantly, and thus cardiac output will not increase. Therefore, it is unlikely to prove beneficial for most women receiving dilute small volumes of local anesthetics for labor epidurals. The main problem with crystalloid studies is likely timing. Crystalloid rapidly moves into the interstitial space and therefore, the increase in central blood volume garnered from an IV bolus of crystalloid (no matter how much) is fleeting. Colloids, because they remain in the intravascular space for a prolonged period, are more effective in studies at both increasing cardiac output and reducing the incidence of hypotension. Unfortunately, colloids are less available, more expensive and have a low risk of severe allergic reaction. Therefore, it may be wise to avoid colloids as part of a per routine policy. One method that has shown promise in the literature is to delay the preload until after the spinal block or concomitant to induction of the spinal anesthetic (Cohydration or coload). Kamenik et al. [1] looked at this using LR as a preload vs. a coload. They found that in the coload group, cardiac output remained elevated above baseline for 30 min after induction of anesthesia. Also, Dyer et al. found that rapid crystalloid infusion during or at the time of spinal induction was more effective at preventing hypotension than a preload of the same volume of crystalloid prior to cesarean section [2]. The same findings were published by Mojica et al. in nonobstetric patients [3]. But, others have found no benefit to coload vs. preload. In particular, one study looked at colloid 500mL either before or after spinal anesthesia and showed no difference. In reality, it is likely that when colloids are used there is no difference. Physiologically this is intuitive given that colloids remain longer in the intravascular space. A recent meta analysis found 5 studies who compared a preload regimen vs. a coload regimen [4]. There meta-analysis was unable to find a benefit to coloading, however there was a trend in the direction favoring coloading. They included studies of both crystalloid and colloid, so perhaps that explains why Dyer et al. and Mojica et al. found a benefit of coloading (used crystalloid) and the metaanalysis did not (included many studies that used colloid).
In summary, mandatory preloading for labor epidural analgesia is not merited. The textbook, "complications in anesthesia" states that no benefit has been shown in regards to reducing the incidence of hypotension or fetal bradycardia by using intravenous preload with crystalloid. The author also makes the point that a liter bolus of crystalloid will often reduce uterine contractions which may lead to unnecessary medical interventions such as starting pitocin or increasing the pitocin dose. In some cases, crystalloid infusions may reduce the severity or increase the responsiveness to vasoconstrictors, but the evidence in favor of this is sparce. In terms of fluid management prior to cesarean section, it would seem that either 1L of colloid given either just prior or with the spinal, or 1L to 1.5L of crystalloid solution with induction of spinal anesthesia may reduce by a small amount the incidence of hypotension. However, no study has ever been able to show that the incidence of hypotension can be made negligible, so vigilance after neuraxial blockade of any type is by far the most important action.

1. Kamenik M, Paver-Erzen V. The effects of lactated ringer's solution infusion on cardiac output changes after spinal anesthesia. A&A 2001;92:710-4.
2. Dyer RA, Farina Z, Joubert IA, Du Toit P, Meyer M, Torr G, Wells K, James MFM. Crystalloid preload versus rapid crystalloid administration after induction of spinal anesthesia (coload) for elective cesarean section. Anaesth Intensive Care 2004;32: 351-7.
3. Mojica JL, Melendez, HJ, Bautista LE. The timing of intravenous crystalloid adminstration and incidence of cardiovascular side effects during spinal anesthesia: the results from a randomized controlled trial. Anesth Analg 2002; 94:432-7.
4. Banerjee et al. Preload or Coload for spinal anesthesia for elective cesarean delivery: a meta-analysis. Can J Anesthesia. 2010;57(1), 37-41.
5. Zamora JE, et al. Haemodynamic consequences and uterine contractions following 0.5 or 1.0 litre crystalloid infusion before obstetric epidural analgesia.Can J Anaesth. 1996;43(4):347-52.

No comments: