Managing Maternal Hypotension During Cesarean Anesthesia

Introduction and Anesthesia Selection

In elective or emergency childbirth, a choice between general and local anesthesia is often required. Because cognitive awareness of the surroundings can benefit the birthing process, a local anesthetic administered via an intrathecal spinal injection or through a catheter in the epidural space offers a distinct advantage. Ratcliffe and Evans at John Radcliffe Hospital in Oxford, England tested this advantage on more than 90 elective cesarean parturients (Ratcliffe & Evans, 1993). The epidural anesthetic group enjoyed the most advantages. The determinants supporting these findings were fetal and maternal health, judged by Apgar scores and the pH values of umbilical blood. The greatest acidities (pH less than 7.2) were observed in neonates in the spinal anesthesia groups. In terms of general fetal health, 70% of neonates from the general anesthesia group did not achieve an Apgar score greater than seven in the first minute after birth.

Apgar scores are associated with the immediate health and viability of the newborn. These scores are tallied at every minute after birth up to five minutes, and up to ten minutes in cases of distress. APGAR is an abbreviation for: Activity (indicative of muscle tone), Pulse, Grimace (a measure of reflex irritability), Appearance, and Respiration (Parer, 1996). A point is awarded if the baby shows active movement, has a pulse over 100 beats per minute, pulls away and cries, breathes and cries, and otherwise appears normal. A score of seven to ten is considered a normal delivery. A score between four and seven necessitates resuscitative action. A score of three or less requires immediate resuscitation, and a score of zero over time is indicative of stillbirth. Apgar scores and other hemodynamic measures serve as indicators of newborn and maternal health. The positioning of the mother can have a significant impact on these measurements.

Blood loss during childbirth, or blood pooling in the extremities following epidurals and local spinal anesthetics — where peripheral arterial resistance is impaired — results in hypotension (Emmett et al., 2002). A meta-analysis was conducted at a Canadian university to assess the efficacies of different treatment modalities for reducing hypotension in women who received spinal and/or epidural anesthetics (Morgan, Halpern, & Tarshis, 2001). Twenty-three studies were considered for this meta-analysis. Efficacies of treatment were determined by reduction of incidences of hypotension and other hemodynamic variables. The effects of volume preloading with crystalloid and colloidal solutions, wrapping with bandages, the use of stockings, and the addition of vasoconstrictors such as ephedrine were all studied. Ephedrine was naturally useful in causing dose-related increases in heart rate and cardiac output — effects needed to counteract hypotension. It was also used as a rescue intervention when maternal hypotension persisted. Results revealed that crystalloid preloading was not effective in reducing incidents of hypotension compared to controls. Bandage wrapping was consistently preferred over stockings. Every study that used colloidal solution preloads indicated that these solutions did help decrease incidents of hypotension, with only one study reporting that colloidal preloading was not useful.

Another meta-analysis of twenty-one relevant studies carried out in Adelaide, Australia indicated that none of the above-mentioned techniques were conclusively effective in reducing hypotension. The meta-analysis compared crystalloid versus colloidal solutions, compression techniques versus controls, and ephedrine versus controls. Besides hypotension, standard hemodynamic variables were measured for the mother and fetus. In all four categories of study, relative risks (odds ratios) were determined as measures of efficacy. The odds ratios for the four study types were 0.78, 0.63, 0.54, and 0.70, respectively (Emmett et al., 2001). Since all values are above 0.50, the results cannot be considered significant under conventional epidemiological standards.

A survey-based study is a good indicator of the current trend in managing maternal hypotension. Some problems with creating experimental conditions stem from fears about risks to neonates (Burns, Cowan, & Wilkes, 2001). Results from different studies are often confounders, and there may be a general tendency to favor established practices. This study was based on a survey of obstetricians and anesthesiologists, and majority opinions are recorded for each technique. The authors reported that Hartmann's solution was overwhelmingly the preload of choice, with most respondents infusing one liter. A left lateral tilt was favored, and ephedrine was the vasopressor of choice for maintaining blood pressure.

One of the problems in identifying the best technique is that results often conflict between studies. There is also a lack of widely accepted guidelines, and researchers frequently use techniques in conjunction with one another — a practice that introduces confounders into the data.

Intravenous Fluid and Volume Loading

Decreased arterial blood pressure in the peripheral regions following local anesthetics is due to loss of vascular resistance caused by the numbing effects, resulting in reduced venous blood flow. Preloading with intravenous infusions is used to maintain preload cardiac volumes, assure stroke rates, and sustain cardiac output. Intravenous infusions attempt to maintain cardiovascular function by mimicking the constituents of blood plasma. These preload solutions can be either crystalloid or colloidal in nature (Vercauteren et al., 1996). The volumes and flow rates of intravenous infusions may also play a role in reducing hypotension, in addition to the compositions of these solutions.

Infusion of Hartmann's solution in 51 patients from a cohort of 104 showed significant improvements in fetal heart rates and maternal hypotension. When compared to the control group of healthy parturients, fetal heart rate abnormalities decreased from 34% to 12%, and cases of maternal hypotension decreased from 28% to 2% (Collins, Bevan, & Beard, 1978).

These results were supported by a subsequent study by Lewis, Thomas, and Wilkes (Lewis, Thomas, & Wilkes, 1983), in which Hartmann's solution was used in 60 patients and deemed a superior preload alternative. Using Hartmann's solution, the incidence of hypotension was reduced to 6.7%.

A study by Kinsella et al. (Kinsella, Lee, & Spencer, 1990) indicated that, although crystalloid solution preloading improved fetal heart rate in test cases versus controls, there were no significant differences in maternal hypotension. This study was conducted at St. Michael's Hospital in Bristol, UK, on 105 women who all received epidural local anesthesia.

A study in South Africa also showed that the flow rate of crystalloid preloading volume did not affect hypotension in parturients. In fact, some members of the rapid infusion cohort suffered from "unacceptable hypertension" (Rout, Akoojee et al., 1992). Twenty patients were divided into two subgroups of ten. Each subgroup received a crystalloid solution of a specific volume over 20 minutes or twice as rapidly over 10 minutes. The number of patients who suffered hypotension was not significantly different between the groups (6 and 7 patients, respectively). The central venous pressure in the rapid infusion group was higher, but this did not impact the incidence of hypotension.

Dextrose is a relatively less commonly used crystalloid solution (Warwick & Weingarten, 1994), and normal saline is not typically indicated because it does not contain a true balance of electrolytes and non-electrolytes. A study of 119 parturients for elective cesarean procedures administered either a 5% solution of dextrose in saline or a normal saline solution, infused two hours prior to delivery (Wilson et al., 1999). Approximately a quarter of a liter was infused intravenously over two hours. Results indicated that adding dextrose to the preload solution did not reduce the incidence of hypotension. The anesthetic cocktail of choice was bupivacaine 0.75%, fentanyl, and morphine, with ephedrine used to treat any resulting hypotension. The need for and dosage of ephedrine did not differ between the dextrose and normal saline groups.

A reevaluation study by Rout and co-workers decisively concluded that there was no need for a mandated fixed volume of crystalloid preload prior to spinal or epidural anesthesia. Their study showed that crystalloid preloading did yield statistically significant reductions in hypotension, but that specific volumes were not necessary. The study included 144 patients — divided unequally between a test subgroup receiving a crystalloid preload 20 minutes prior to anesthetic injection and a control subgroup receiving no preload. Seventy-one percent of the control group suffered from hypotension versus 55% of the test subgroup. Hypotension was defined as a systolic arterial pressure of 100 mmHg or a 20% decrease from baseline SAP (Rout et al., 1993).

Researchers at the Queen Mother Hospital in Glasgow, Scotland also abandoned the routine of specific volume preloading. They studied 60 healthy elective cesarean delivery women using 1000 mL versus 200 mL of crystalloid preload (Jackson, Reid, & Thorburn, 1995). Ephedrine was administered as soon as hypotension was observed, with the threshold set at 90 mmHg or 30% of baseline pressure. The significant difference in preload volumes did not significantly affect the incidence of hypotension. Ten women in the 1000 mL cohort and nine in the 200 mL cohort experienced hypotension lasting longer than three minutes before ephedrine was administered.

While those two studies indicate that crystalloid volume has little effect on overall hemodynamic outcome, a study by Hahn and Resby at the Karolinska Institute in Sweden reported that the body processes volume differently in attempting to maximize venous return preload, and that this mechanism may affect maternal hypotension. Volume handling was studied using lactated Ringer's solution and a 3% dextran solution, with blood hemoglobin concentrations and urine output as parameters (Hahn & Resby, 1998). No conclusive mechanism was identified for how the body compensates for volume loss with infused solutions, as the observed results must be applied to theoretical models whose outputs vary.

A study conducted at Harvard Medical School further questioned the necessity of maintaining crystalloid volume. Results for 55 parturient women administered 10, 20, and 30 mL/kg of crystalloid preload solutions indicated that neither the onset nor the duration of hypotension changed across the three volumes (Park et al., 1996). The need for ephedrine intervention did not differ among the three groups, and the researchers also cautioned that larger preload volumes carry a significant risk of maternal or fetal edema.

A newer school of thought proposes precluding preload solutions by administering lower doses of anesthetic and epidural blocks, on the theory that lower concentrations will enhance venous returns and reduce hypotension. Hofmeyr conducted a meta-analysis to evaluate this approach and acknowledged significant potential biases in the selected clinical trials (Hofmeyr, 2002). One study indicating no increase in maternal hypotension or abnormal fetal hemodynamics with low anesthetic doses was noted, though the cohort was too small to draw firm conclusions.

The use of colloids in preventing post-anesthetic hypotension has been shown in multiple studies to have more beneficial effects than crystalloid solutions alone. A study by Kee et al. from the Chinese University of Hong Kong on 68 patients revealed that a colloidal preload largely prevented hypotension. Thirty-three parturients received a 4% gelatin solution (Gelofusine) at 15 mL/kg, while a control group of 35 received no colloidal preload (Ngan Kee et al., 2001). The control group required more rapid infusion of the vasopressor metaraminol at greater concentrations — on average 0.6 mg/mL more — and had a lower mean SAP nadir by approximately 7 mmHg.

Several direct-comparison studies confirm that colloid preloading is more beneficial than crystalloid preloading and that colloids offer the advantage of requiring smaller volumes. A study of 26 healthy parturients undergoing elective cesarean sections given either 500 mL of lactated Ringer's solution or half that volume of a hydroxyethyl starch colloidal solution found that the incidence of maternal hypotension was significantly larger in the crystalloid group (62% versus 38%). Fetal hemodynamics based on Apgar scores did not differ between groups. Researchers at the Oulu University Central Hospital in Finland agreed the results were inconclusive, noting that the pulsatility index — a measure of uterine arterial response — spiked and recovered quickly regardless of subgroup, suggesting uterine arterial response does not depend on the preload solution (Karinen et al., 1994).

In a follow-up study by the same Finnish group, fetal pulsatility indices and hemodynamics were examined using a similar cohort and the same 2:1 crystalloid-to-colloid volume ratio. Results indicated that colloid versus crystalloid preloads had no effect on birth outcome or neonatal health. Minor additional observations included a general decrease in maternal pulsatility index with colloidal infusion, and a temporary increase in fetal heart rate with crystalloid infusion (Karinen et al., 1995). A study at the Hammersmith Hospital in London of 50 women with elective cesarean sections found no significant hemodynamic differences between two liters of crystalloid preload and a combination of two liters crystalloid plus one liter colloid.

A study of 20 women given either Hartmann's solution alone (1 liter) or a combination of Hartmann's solution with 0.5 liters of 5% polygelatin (Haemaccel) showed that the mixture significantly reduced hypotension — 5% in the mixture group versus 45% in the crystalloid-only group (Murray, Morgan, & Whitwam, 1989).

A study by Sharma and co-workers revealed that a 6% colloidal solution (500 mL) was significantly more effective at reducing post-spinal hypotension for tubal ligations than twice the volume of lactated Ringer's solution (Sharma, Gajraj, & Sidawi, 1997). Of 40 participants, 52% of the 21 patients in the Ringer's group required intervention for hypotension, compared with only 16% of the 19 patients in the colloid group. More ephedrine was also required in the crystalloid group.

Similarly overwhelming benefits for colloids were not observed in a study of elderly patients undergoing hip replacement surgery. While the colloid group (Haemaccel, 500 mL) showed higher overall systolic blood pressure, morbidity from hypotension was not significantly different across the groups receiving Hartmann's crystalloid, colloid, or no pre-hydration (Hallworth, Jellicoe, & Wilkes, 1982). There were also no differences in ephedrine dosage requirements for cases of severe or mild hypotension.

A study by Riley and co-workers came out strongly in favor of colloid preloading over crystalloid. Forty women opting for cesarean delivery were divided into two groups: one receiving 2 liters of lactated Ringer's solution and one receiving 500 mL of 6% hetastarch plus 1 liter of Ringer's solution (Riley et al., 1995). Over 85% of the lactated Ringer's group showed symptoms of hypotension versus 45% of the hetastarch group. The hetastarch group had a higher mean SAP by 7 mmHg. The Ringer's group also had a higher heart rate, a shorter time to onset of hypotension, and required more frequent and higher doses of ephedrine. The researchers recommend hetastarch as a standard preload prior to local spinal or epidural anesthetics.

Regarding serum albumin, a study of 60 patients divided into three groups — each receiving 1200 mL, either exclusively as lactated Ringer's or as various Ringer's/albumin combinations — showed that all groups fared similarly, with no advantage to any group (Ramanathan et al., 1983). Albumin levels measured immediately and 24 hours after infusion did not differ meaningfully.

Researchers at the University of Gothenburg in Sweden tested whether smaller volumes of colloid solution mixed with prophylactic concentrations of ephedrine could replace large-volume preloads. Two groups received either a mixture of 7.5 mg of ephedrine with 3% dextran 70, or twice the volume of dextran solution alone (Wennberg et al., 1992). There were no advantages in reducing hypotension between the two regimens, indicating that larger volumes were not necessary when a mixture would work equally well. The researchers also noted lower incidences of maternal nausea in the mixture group.

Vasopressors: Ephedrine

One of the problems with identifying the appropriate anesthetic dosage is that too little results in discomfort from peripheral pain, while too much causes maternal hypotension due to impaired venous return. The anesthetist must balance these conditions to achieve optimal outcomes for the obstetric procedure. In most clinical trials, ephedrine is identified as the gold standard for immediate rescue: it corrects decreases in blood pressure by increasing cardiac output, heart rate, and arterial pressure through a combination of alpha- and beta-adrenoreceptor mechanisms. Because of its positive effects in reducing hypotension, ephedrine may also be considered as a prophylactic agent. Clinical trials are then necessary to identify the appropriate dosage, given the danger of hypertension and tachycardia at higher doses, which can endanger fetal life and introduce neonatal abnormalities.

In parturients prone to hypertension, alternative vasopressors must also be considered. This section explores the role of ephedrine as both an intervention and a prophylactic, as well as the role of phenylephrine and combined regimens.

In order to identify the appropriate dose of ephedrine, Simon et al. at the Hôpital Saint-Vincent de Paul in Paris divided 108 women opting for cesarean delivery into three subgroups (Simon et al., 2001). Groups received 10 mg, 15 mg, or 20 mg of ephedrine intravenously, administered two minutes after spinal anesthesia. Patients were then uniformly placed in a supine position with a 15-degree left lateral tilt. Mean arterial pressures and systolic and diastolic pressures were measured, with 30% above or below baseline as the threshold. Boluses of 5 mg ephedrine were given to correct hypotension, with a maximum of 50 mg for any patient. The study concluded that 10 mg was insufficient to combat hypotension. Better results were obtained with 15 and 20 mg doses. Though seven of the 108 women showed signs of hypertension after ephedrine infusion (two, two, and three from the three subgroups, respectively), the distribution did not enable a correlation between hypertension and ephedrine dose strength.

Ephedrine has also been shown to work as an oral prophylactic. Clinical trials on 100 women at a hospital in Nepal, in which spinal local anesthetics were administered for abdominal surgery, found that a prior oral dose of 30 mg of ephedrine significantly reduced presentations of hypotension compared to a control group (Kafle, Malla, & Lekhak, 1994). The need for supplemental ephedrine was more than twice as high in the control group, as was the need for inotrope fluid treatment.

The study by Simon et al. cites a study by Kee and co-workers at the Chinese University of Hong Kong, which identified 30 mg of IV ephedrine as effective in reducing maternal post-anesthetic hypotension (Kee et al., 2000). The French group criticized this finding, noting that the high dose produced unacceptable levels of hypertension (Simon et al., 2001). In the Hong Kong study, the smallest average incidence of hypotension was found in the 30 mg group, with the 10 mg and 20 mg groups showing statistically similar — and higher — hypotension rates. The Simon group averred that while 10 mg is less than optimal, 30 mg risks hypertension, and 20 mg is optimal for reducing maternal hypotension. This study was conducted for 80 women undergoing elective cesarean delivery. The Hong Kong authors conceded that significant cases of hypertension occurred even in lower concentration groups, including one hypertensive patient in the control group.

Researchers Chan et al., following a study from the Queen Mary Hospital in Hong Kong, concluded that prophylactic doses of ephedrine prior to anesthesia can preclude the need for crystalloid preloading. Forty-six women undergoing elective cesarean procedures were given either a Hartmann's solution preload or 0.25 mg/kg of ephedrine (Chan et al., 1997). Severe hypotension (a decline of more than 30% from baseline SAP) occurred almost twice as often in patients receiving the crystalloid preload (65% versus 35%). Shivering was more than three times more common in the Hartmann's group, and better umbilical pH values were observed in the ephedrine group.

In a study from Holbaek Central Hospital in Denmark, 48 patients undergoing spinal anesthesia were divided equally into three groups. All received a preload; the first group received 12.5 mg of ephedrine intravenously, the second group received 37.5 mg intramuscularly, and the placebo group received no ephedrine (Hemmingsen, Poulsen, & Risbo, 1989). Hypotension was significantly more common in the placebo group. The researchers also assessed outcomes by pre-procedure ASA (American Society of Anesthesiologists) classification, noting that higher-risk patients were more likely to suffer from hypotension — some falling more than 33% below baseline.

At the University Hospital in Antwerp, Belgium, the relative efficacy of 5 mg of ephedrine versus a placebo (normal saline) was tested in parturients. The saline or ephedrine was given immediately after spinal anesthesia, with all patients preloaded with a combination of 1 liter of Hartmann's solution and 500 mL of hetastarch. Fifty-eight percent of the saline patients presented with hypotension versus 25% of the ephedrine group, and only two members of the ephedrine group had severe hypotension compared to ten from the saline group (Kee et al., 2000). The Belgian researchers concluded that this result holds for prehydrated parturients, and noted that the colloidal component of the preload mixture may also have contributed to the reduction in hypotension.

Dosage remains a critical concern. Rout et al. from the University of Natal in South Africa concluded that intramuscular administration of high doses of ephedrine should not be recommended, and that the use of 50 mg doses should be abandoned (Rout, Rocke et al., 1992). Their study of 30 healthy parturients assigned to 25 mg, 50 mg, or saline control showed hypertension in nine patients in the 50 mg group and five in the 25 mg group, with average increases in pressure exceeding 28% over baseline.

Beneficial results were observed with lower doses in clinical trials at the Rotunda Hospital in Dublin, Ireland. A cohort of 68 parturients receiving a lactated Ringer's preload were given 6 mg or 12 mg of ephedrine, or a 0.9% NaCl control, post-anesthesia (Loughrey, Walsh, & Gardiner, 2002). The 12 mg administration was found to be ideal in preventing hypotension, and ephedrine intervention needs were also significantly lower in this group.

Fong et al. at New York-Cornell Hospital tested whether ephedrine sulfate could be injected as part of the epidural cocktail (Fong et al., 1996). Patients were preloaded with a colloidal solution and placed in a supine position with a right wedge and 15-degree tilt, with hypotension defined as 90 mmHg or 70% of baseline. Results were inconclusive, and the test group actually presented with more hypotension, though differences were not statistically significant. The researchers note that ephedrine sulfate injected into the epidural space likely does not access the bloodstream rapidly enough to increase heart rate, cardiac output, and blood pressure effectively. Similarly, Tsen et al. at Harvard Medical School observed no improvements when ephedrine was mixed with the spinal anesthetic; both test and control groups presented with maternal hypotension 70% of the time (Tsen et al., 2000).

In a departure from the generally positive findings for ephedrine, a study of 122 healthy parturients given a 10 mg bolus of ephedrine versus a control showed no difference in the incidence of hypotension between groups (Shearer et al., 1996). Hemodynamic studies on the fetus revealed that ephedrine in this case resulted in lower umbilical pH values.