Impact of adding midazolam to bupivacaine 0.5% in regional spinal anesthesia on maternal middle cerebral artery velocimetry in parturients with severe preclampsia

Severe preclampsia is a challenging issue facing both intensivist and anesthetic team carrying both maternal and fetal morbidity and mortality. Termination of pregnancy after blood pressure control is the golden key in management. Cerebral complications due to diffuse cerebral vasospasm are most common and serious. Intrathecal midazolam with its gamma amino butyric action may antidote glutamate mediated sympathetic surge and decreasing cererbral vasospasm. Temporal view transcranial doppler imaging maternal middle cerebral artery is used to examine Blood ow indices namely pulsatality index and resisitive index. One hundered Ladies with severe preclampsia scheduled for urgent caeserian section were recruited in 2 groups, both received 10mg bupivacaine 0.5%, Midazolam group received 1mg midazolam and the other group received 0.2ml sterile saline 0.9% NaCl. All vascular indices were signicantly better in midazolam group, less ICU stay. and 48 hours postpartum. They determined that these patients had elevated systolic, diastolic, and mean velocities in the MCA 24 hours postpartum compared with those in the antenatal period. They also demonstrated a further increase in all velocities at 48 hours postpartum. The same investigators also found that preeclamptic women, compared with normotensive women, had elevated cerebral blood ow velocities in the antepartum period and at 24 and 48 hours postpartum. These data suggest that the cerebral vasculature of preeclamptic women is in a vasoconstricted state before delivery as well as in the immediate postpartum period (i.e., up to 48 hours after delivery) compared with normotensive women.


Background
Preeclampsia is a disorder of widespread vascular endothelial malfunction and vasospasm occurring after 20 weeks of pregnancy and can present as late as 4-6 weeks postpartum [1]. The incidence of preeclampsia in the United States is estimated to range from 2% to 6% in healthy, nulliparous women [2]. In developing nations, the incidence of the disease is reported to be 4-18% [3], with hypertensive disorders being the second most common obstetric cause of stillbirths and early neonatal deaths in these countries [4].
Risk factors for preeclampsia include primigravida, diabetes mellitus, kidney disease, chronic hypertension, previous personal or family history of pre-eclampsia, maternal age more than 35 years, obesity, antiphospholipid antibody syndrome and multiple gestations [5].
Pre-eclampsia can be diagnosed through Systolic blood pressure ≥140 mm Hg or diastolic ≥90 mm Hg on two separate readings taken at least four to six hours apart after twenty weeks [6].In a woman with essential hypertension beginning before twenty weeks gestational age, the diagnostic criteria include an increase in systolic blood pressure (SBP) by ≥ 30mmHg or an increase in diastolic blood pressure (DBP) ≥15 mmHg, also Proteinuria ≥ 3 grams (300mg) is considered a diagnostic tool [7].
The rationale for the administration of intrathecal midazolam focuses on the awareness that it is an agonist at the benzodiazepine binding site, a subunit of the pentameric gamma-aminobutyric acid (GABA-A) receptor. The agonist occupancy of the benzodiazepine binding site enhances the activity of GABA at the GABA-A receptor [8].
Transcranial Doppler ultrasonography (TCD) is considered the sole noninvasive real-time neuroimaging modality for the evaluation of characteristics of blood ow in basal intracerebral vessels. TCD adds physiologic information to structural imaging. Additionally, TCD has been rapidly evolving from a simple, noninvasive diagnostic tool to an imaging modality with a broad spectrum of clinical applications [9].

Methods
This prospective randomized double-blind study was approved by the local ethics committee of the Faculty of Medicine, ElMinia University, and registered at clinical trials under the number of (NCT04283110). It was adhered to the Declaration of Helsinki and involved 100 ladies recruited into two groups, 50 lady each . Both groups received subarachinoid block with 10 mg bupivacaine 0.5% plus 1mg of midazolam in midazolam group or 0.2 ml sterile saline 0.9% NaCl as a placebo for equi-volume injection. Exclusion criteria included refusal for regional anesthesia or any condition prohibit subarachinoid block as coagulopathy.
An informed consent has been obtained from all participants who were randomly and equally allocated into two groups through web-based randomizer (https://www.randomizer.org/). Double blind fashion was executed ( neither the obsever nor the parturients was aware of study design) Diagnosis of severe preclamsia was held by senior obstetric resident of charge at the emergency room (ER). Two cardinal criteria must be checked, noninvasive blood pressure (NIBP) ≥140/90 mmHg (two readings 5 minutes apart), albuminuria > +++ by boiling method. Presentation by imminent symptoms as headache, blurring of vision and nausea, HELLP syndrome ( hemolysis of RBCs, elevated liver enzymes and low platelets) (hemolysis of RBC evidenced by elevated indirect bilirubin and lactate dehydrogenase > 600 IU, platelet count < 100.000, if two of the above mentioned criteria were found, diagnosis of severe preeclampsia was declared. Pelvic ultrasound was done to con rm fetal state to exclude any fetal compromise and clarify complications. Immediately after diagnosis, admission to obstetric Intensive care unit , Two wide bore cannulae (20 gauge, ultramed, Belguim) were inserted, one for drugs and the other for emergency. Full monitoring tools were applied (NIBP, Electrocardiography, pulse oximetry and urinary foley catheter (16 gauge) was inserted after vaginal sterilization with Povidone Iodine (7.5%). Investigations were performed: complete blood picture, liver functions , renal functions (serum urea and serum creatinine). Loading dose of Magnesium Sulphate (MgSO 4 ) (1gm/10ml amp) 4gm on 200 ml Ringer lactate over 20 min followed by a maintenance infusion of 1 to 2 g/h by controlled infusion pump. Our target was to control MAP between 90-100mmHg over 45 minutes, during MgO4 infusion, Meticulous monitoring for signs of MgSO4 toxicity (respiratory rate <16 cycles per min, loss of knee jerk re ex, urine output less than 0.5 ml/kg/h). Any sign of the above mentioned criteria mandated immediate stop infusion of MgSO4 at once and serum Magnesuim was checked. If MAP was ≥110mmHg despite of MgSO4 infusion, Nitroglycerin (Nitronal, 50mg /50ml,Sunny pharmaceutical) in an incremental dose starting with 1 mic/kg/h , upper maximum dose was 10 mic/kg. Our goal was to control of MAP gently to 80-100mmHg. Before C.S, another pelvic U/S was performed to evaluate fetal state. Brain imaging (brain CT and MCA trans-cranial Doppler performed and the results were documented. Throughout this management, All resuscitative measures (endotracheal tube, Atropine and Adrenaline) were ready for any emergency beside continuous cardiotocography (CTG) to detect any pathological fetal deceleration. Transcranial Doppler probe (Toshiba aplio500 with convex probe 2-5mhz) rested on the temporal window (just above the zygomatic arch between eye &ear). Ladies were positioned in a semi-recumbent position with a 15-degree left lateral tilt.
A transcranial Doppler probe with a 10 mm sample volume was used to insonate the M1 ( rst 2 cm of the middle cerebral artery) portion of the middle cerebral artery by means of the transtemporal approach then the mean ow velocity, pulsitility index and resistance index values were recorded.
Data collection included haemodynamic data, urine output, serum urea and creatinine, dose and duration of nitroglycerine, pulsatality index, resistive index, mean ow velocity and incidence of complications.
Statistical analysis: We have determined a signi cant difference in the values of pulsatality index and resistive index by using the power of 80% and a signi cance level of 5%, and accordingly sample size was determined to be 50 participants in each group. Collected data were rstly assessed by Kolmogorov-Smirnov for the normality, and were presented as a number, percentage, mean ± SD or median (range). Chi-square test and Fisher Exact test were used to compare qualitative variables. Continuous variables were compared with t-test (Parametric data) or Mann-Whitney U test (Non-parametric data). A two-tailed p < 0.05 was considered statistically signi cant. P-value considered statistically signi cant when P < 0.05.
Data entry and data analysis were done using SPSS version 19 (Statistical Package for Social Science.

Results
Parameters assessed : Table 1 shows the demographic data (age) of the parturients. The 100 participants were equally randomized between the two groups as shown in the CONSORT ow-chart ( Fig.1), and they were comparable regarding their demographic details with insigni cant differences in between. Table 2 showed Changes in systolic blood pressure (SBP) in studied groups: Regarding changes in SBP, there was a statistically signi cant difference between the two groups immediately after induction, 5 min, 10 min after induction, after delivery of placenta, 20 min, 30 min of operation and all values during rst 24 h postoperativly. Intragroup comparison showed signi cant statistical difference between pre induction values and all intra and postoperative values in both groups. Decline in SBP was steady and gentle in (M) group while it was sharp in (C) group. Rebound systolic hypertension was more evident in (C) group rather than (M) group. Table 3 shows Changes in diastolic blood pressure (mmHg) in studied groups: Regarding changes in DBP among the studied groups pre ,intra and post operatively were statistically signi cant difference between the two groups, 5 min, 10 min after induction, after delivery of placenta, one h, four hs, six hs, 12 hs, 18 hs and 24 h postoperatively.
Intragroup comparison showed signi cant statistically difference sbetween pre induction values and all intra and postoperative values in both groups.
DBP decreased gently in (M) group but it showed sharp decrease in the (C) group with noticeable rebound hypertension occurred postoperatively. Table 4 of nitroglycerine dose, Post operatively, 24 cases in (C) group mandated use of nitroglycerine to control postoperative rise in blood pressure with mean±SD 3024±1855.1 mic/kg and total duration 754 hs (32 days nearly) while in (M) group, statistically less number of cases (7) and less incremental dose( mean±SD 2013±1210.4 mic) and short duration (174 hours, about 8 days). There was a signi cant difference between the studied groups regarding (Number of cases, total dose & total duration of nitroglycerine infusion with (p value = 0.004)) Table (5) shows mean serum creatinine among studied groups. With no statistically signi cant difference between studied groups as regard serum creatinine . Table 6 shows fetal APFAR score. There is no signi cant difference among studied groups regarding APGAR score.
A: activity, P: pulse, G: grimace, A: appearance, R: respiration. Table 7 represents complications among studied groups: Table 8 represents pulsatality index among studied groups, There was a signi cant difference between studied groups at 6 h and 24 h postoperatively with (p value <0.001 , <0.001) for the Rt side and (0.003 and , 0.004) for the Lt side.
There was also a signi cant difference between the PI values among (M) group at 6h and 24 h compared with basal value on both sides. Table 9 of MFV among studied groups, There was no signi cant difference in between or inside studied groups by comparing the MFV.

Discussion
Preeclampsia is a challenging obestetric emergency facing both anesthetists and intensivists. Preeclampsia confers high incidence of maternal and fetal morbidity. During the period of the current study, the incidence of severe preeclampsia in our hospital was 15.4% (352 preeclamptic ladies out of 2282 presented in the ER for emergency termination of pregnancy). Severe preeclampsia is the most common cause of urgent C.S and admission to obstetric ICU in our center The middle cerebral artery is the largest branch of the internal carotid artery carrying 75% of total CBF supplies a portion of the frontal lobe and the lateral surface of the temporal and parietal lobes. Transcranial doppler is the only noninvasive real-time neuroimaging modality for the evaluation of characteristics of blood ow in intracrnial vessels that adds physiologic information to structural imaging. TCD can provide information about vascular stenosis and occlusion, the hemodynamic status of the cerebral circulation, and real-time monitoring of vascular indices. TCD is useful for detecting increased intracranial pressure (Tsivgoulisetal.,2009).
In the current study, TCD assess the proximal portion of the MCA through temporal window, this examined segment contains two types of receptors; Serotonin (5HT) and Dopamine receptors( D) that are not affected by the acetylcholine, so free from effect of MgSO 4 on these receptors (Purkayastha and Sorond, 2012). This clari es any change in the values of resistive indices is exclusively explained by intrathecal midazolam effect.
Control (C) Group showed a noticeable decline in both systolic and diastolic blood pressure either after subarachnoid block or after placental delivery while postoperatively, there was a rebound systolic and diastolic hypertension in the same group. Consequently, more number of cases (24) were in need to higher incremental doses of nitroglycerine (mean ±SD (3024±1855.01mic/kg)) and lasting for longer duration (32 days). In (M) group, we noticed that decline was steady and gentle in both systolic and diastolic blood pressure after subarachnoid block and placental delivery with minimal rebound hypertension postoperativly, less number of ladies (7) were in need of incremental doses of nitroglycerine (mean±SD2013±1210.4) with less duration (8 days explained by the fact that midazolam has a bupivacaine -sparing effect and thus causes gradual sympathectomy. Pharrmcokinetics of midazolam points to its preferential diffusion to lamina II in posterior horn cell in grey matter of spinal cord away from the lateral horn cells hosting sympathetic nuclei thus preserve sympathetic tone to some extent (Shadangi etal,2011) In agreement of our study, Sanwaletal,2013 studied the bupivacaine sparing effect of intrathecal midazolam in sub-arachnoid block for cesarean section and discovered that the hypotensive episodes was greater in control group while all Midazolam groups ( 1mg, 1.5mg and 2 mg) showed steep values of SBP and DBP( less episodes of hypotension).
Neuronal complications of severe preeclampsia (eclampsia, PRESS and intracranial hemorrhage) are the most common cause of maternal and fetal morbidity and mortality. Incidence of neuronal complications was higher in (C)group, 40cases) while only 9 cases in (M)group.
Intracrnial haemorrhage was in focus in one lady in (C) group. She experienced both systolic and diastolic hypertension, early nitroglycerin infusion started immediately after ICU admission for 24 hours postpartum. The lady developed 2 attacks of eclamptic ts during this period. Mechanical ventilation was initited immidiatelly after rst t to protect airway and prevention of aspiration pneumonia. Brain CT revealed massive lt parieto-occipital hematoma.
Acute kidney injury in sever preeclampsia is the second cause for maternal and fetal morbidity and mortality. Diffuse release of vasospastic materials (endothelin, serotonin, noradrenaline and abruptio placenta) are the main stay for both direct and indirect renal injury.
Urine output, serum urea and serum creatinine were higher in (C) group, while in (M) group it showed less amount of urine output with lower serum urea and creatinine. This can be explained as there were 2 cases in (C) group with acute kidney injury managed by forced diuresis. (M) group showed signi cant lower incidence than(C) group. For renal impairment at (C) group; one case (2%) had acute renal injury with subsequent elevation in the renal function responded to uid challenge and forced diuresis. Also there was a case of acute renal failure was refractory to uid challenge and frusemide infusion, was elicited for hemodialysis.
In agreement with our study, Goplani et al, 2008 reported a study between January 2004 and May 2006, 772 patients with ARF were admitted at the Institute Of Kidney Diseases and Research Centre and Institute Of Transplantation Sciences. A total of 92 patients with ARF were pregnant. Of these, 22 patients had the evidence of renal disease prior to pregnancy and were excluded; hence, 70 patients with pregnancy-related ARF were studied.
Pregnant women who were healthy previously and had developed ARF were diagnosed in oliguria (Urine output <400 ml/d) and for mounting azotemia (Serum creatinine >2mg%). The incidence of ARF due to pregnancy related hypertension was 10%. A majority of the patients (97.14%) underwent hemodialysis, while one died without dialysis; further, one patient underwent peritoneal dialysis due to hypotension and died later. Peritoneal dialysis was initially performed in 8 patients (11.42%) due to initial hypotension, 7 patients among which improved and received hemodialysis later.
Impaction of sever preeclampsia on liver is evident. Increased hepatic vascular bed resistance, We noticed that serum ALT post operatively 24 h was signi cant in control group than in M group. ALT is more sensitivein cases of acute liver injury. This nding can be explained by more number of HELLP syndrome parturient preoperatively in control group. For liver impairment, in (C) group one case experienced acute fulminant hepatitis 24 h postpartum. This case was included in HELLP syndrome criteria (elevated ALT, AST more than 5 folds, Platelets<100000) secondary to vasospasm of the hepatic vascular bed. Patient responded to medical management (liver support, Vit k) patient was referred to tropical department with no mortality.
Regarding non cardiogenic pulmonary edema, two cases at (C) group were detected after 12 h postpartum,managed with forced diuresis(frusimide 0.2-2 mg/kg/h). Two patient were mechanical ventilated after failure of force diuresis (RR 35 c/m, PH < 7.1 and Spo2< 90%). Both patients could not be weaned from mechanical ventilation, both parturient were mechanically ventilated for 3 days. At (M) group one case with non cardiogenic pulmonary edemamanaged with mechanical ventilation and successfully weaned after 24 h.In the all above mentioned cases echo cardiograph was done revealed normal LT ventricular systolic function.
Mortality rate was higher in (C) group than (M) group. Three cases were enrolled in mortality, 2 parturients with non cardiogenic pulmonary edema and on case with ICH.No mortality was detected in (M) group.
Transcranial Doppler was very helpful in detection and prediction of incidence ofeclampsia as there were a noticeable elevation in MCA resistive indices in (C)group rather than (M) group. Cerebral vasospasm as a component of the pathophysiology in eclamptic women has been consistently described with transcranial Doppler ultrasound. This is illustrated by elevated middle cerebral blood ow velocities and high resistive and pulsatility indices in patients with preeclampsia. Doppler indices in MCA has been shown to be signi cantly higher in preeclamptic women in control group as compared with that in preeclamptic women in midazolam group.Many of the agents used have potent vascular effects and may have caused cerebral vasodilatation. Reduction in the MCA pulsatility index after intrathecal midazolam therapy suggests that vasospasm in MCA is alleviated.
Regarding PI and RI they were of great value in both correlation and prediction of neuronal complications (PRESS, eclampsia). Intrathecl midazolam has dampened the PI value post partum 6 h and 24h, this can be attributed to antidotingof GABA mimetic action of intrathecal midazolam against excitatory glutamate. Strong correlation with ischemic infarction and post partum eclampsia was conclusive speci cally in PI and RI as they expressed the total vascular resistance of MCA to blood ow.
We reported high predictive value of PI (>1.3) with incidence of eclampsia reached to >80%. PI changed more than RI with relief of excitatory glutamate and aspartate. PI was more sensitive than RI in both prediction and correlation of neuronal complications.
Risk strati cation based on serial management by TCD indices concluded that patients withPI (1-1.3) at risk of PRESS, PI (1.3-1.8) at risk of eclampsia and PI (1.8-2.5) at risk of ICH.
In controversy with us, Riskin et al, 2002studied 166 women in the secondtrimester of pregnancy to measure peak, end-diastolic, and mean velocities in the middle cerebral arteries.Middlecerebral arteries pulsatility and resistive indices were lower in the women with preeclampsia who were initiallynormotensive compared with the pregnant women who were normotensive (0.83 and 0.54 vs 0.73 and 0.50, respectively; P < .05). This can be explained by early use of oral antihypertensive drugs (alphamethyldopa) which mediate cerebral vascular vasodilatation.
Preeclamptic patients in both groupshave been evaluated with transcranial Doppler ultrasound in the immediate and extended postpartum periods. Williams et al,2015 investigated 46 preeclamptic women in the antepartum period and again at 24 and 48 hours postpartum. They determined that these patients had elevated systolic, diastolic, and mean velocities in the MCA 24 hours postpartum compared with those in the antenatal period. They also demonstrated a further increase in all velocities at 48 hours postpartum. The same investigators also found that preeclamptic women, compared with normotensive women, had elevated cerebral blood ow velocities in the antepartum period and at 24 and 48 hours postpartum. These data suggest that the cerebral vasculature of preeclamptic women is in a vasoconstricted state before delivery as well as in the immediate postpartum period (i.e., up to 48 hours after delivery) compared with normotensive women.
TCD velocimetry indices were very helpful in prediction and detection of incidence of complications , when regression analysis was done for PI basal, PI 6 hit showed that the most powerful index that can predict complications was PI (R 2 = 0.405). This means that two indices included can predict 81% of cases with complications which is highly signi cant value.

Ethics declarations
Ethics approval and consent to participate The current research is adherent to declaration of Helsinki over ethical committee record 293.9/2019. Written consent was taken from lady herself or next of kinn , Consent is in arabic and including agreement for ceserian section and anther one for research participation.

Consent for publication
Not applicable.    Table 7 representing fetal APGAR score.

Midazolam group N=50
Control group N=50

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.