Addition of Neostigmine and Atropine to Conventional Management of Postdural Puncture Headache

A Randomized Controlled Trial

Ahmed Abdelaal Ahmed Mahmoud, MD, FCAI; Amr Zaki Mansour, MD; Hany Mahmoud Yassin, MD; Hazem Abdelwahab Hussein, MD; Ahmed Moustafa Kamal, MD; Mohamed Elayashy, MD, FCAI; Mohamed Farid Elemady, MD; Hany W. Elkady, MD; Hatem Elmoutaz Mahmoud, MD; Barbara Cusack, LRCP&SI, MB MCh, NUI, MCAI; Hisham Hosny, MD; Mohamed Abdelhaq, MD


Anesth Analg. 2018;127(6):1434-1439. 

In This Article


This prospective, randomized, controlled, double-blind trial was performed after approval by the Research and Ethics Committee of Faculty of Medicine at Beni-Suef University on March 25, 2014. Written informed consent was obtained from all trial participants, and the study period extended from October 15, 2015 to September 8, 2017. The trial was registered at the Pan African Clinical Trial Registry (, PACTR201510001299332) on October 9, 2015 with Ahmed Abdelaal Ahmed Mahmoud as the principal investigator. The trial was registered after completing a pilot study of 20 patients to calculate the sample size of this trial and before enrollment of the first patient. The pilot study was performed from March 26, 2014 to June 3, 2015, and the patient data were not included in this trial.

As noted below, 90 patients of 20 to 40 years of age with American Society of Anesthesiology physical status II because of pregnancy and diagnosed with PDPH following intrathecal spinal anesthesia for elective cesarean delivery were included. Diagnosis of PDPH was based on the International Headache Society criteria ( Patients with PDPH and a visual analog scale (VAS) score <5, a history of chronic headache, cluster headache, migraine, convulsions, cerebrovascular accident, signs of meningismus, preeclampsia, eclampsia, coagulopathy, previous neurological diseases, and severe bleeding (>20% of blood volume); undergoing treatment with vasopressors, bronchial asthma, arrhythmia, and any type of heart block; weighing <50 kg; and with any contraindication of oral intake were excluded.

Intrathecal spinal anesthesia was performed after giving an intravenous (IV) fluid preload with 10 mL/kg Ringer's lactate by an anesthesiologist not involved in the trial. Intrathecal blocks were performed in the seated position using 2.5-mL hyperbaric 0.5% bupivacaine (12.5 mg) at L3–L4 using a 22-gauge Quincke spinal needle (B. Braun, Melsungen, Germany), which is the smallest available spinal needle in our institution because of cost and availability considerations. Parturients with postoperative PDPH and a VAS score of ≥5 were randomly allocated to receive either slow IV injection of 20 μg/kg neostigmine and 10 μg/kg atropine in 20 mL of 0.9% saline given over 5 minutes every 8 hours (n = 41) or 20 mL of 0.9% saline IV every 8 hours (n = 44). The intervention was continued until achieving a VAS score ≤3 or for a maximum of 72 hours. Patient in the neostigmine group who achieved VAS scores ≤3 before 72 hours were given 20 ml of saline 0.9% IV every 8 hours to maintain blinding. Both groups received conservative management, which consisted of nursing in the supine position, hydration with continuous infusion of 30 mL/kg/day Ringer's lactate solution, 1 g paracetamol plus 135 mg caffeine every 6 hours. Ketoprofen (100 mg) suppositories were given twice daily for 5 days as a part of a routine postoperative pain management protocol.

Randomization was performed using sealed opaque envelopes that contained random numbers generated by online application ( The study was double blinded. Participants were not aware of their group assignment, and the medications were prepared by an anesthetist who was not involved in the trial. The anesthetist who assessed the participants after the intervention was blinded to the group allocation. Following World Health Organization recommendations, participants were instructed to withhold breastfeeding for 24 hours after the last dose of neostigmine/atropine.[18] A breast pump was used to relieve breast engorgement, ie, pump and dump. Participants were asked to report the severity of their headache after sitting upright for 15 minutes, using a 10-cm VAS at 0, 6, 12, 24, 36, 48, and 72 hours. The presence of neck stiffness, nausea and vomiting, diarrhea, abdominal cramps, muscle cramps, muscle twitches, bronchospasm, and urinary bladder hyperactivity were recorded throughout the study period as yes or no. Participant age, weight, height, body mass index, and the time interval between dural puncture and the occurrence of PDPH were also recorded. Conservative management of PDPH using oral medications continued throughout. An EBP was performed during the study if the VAS was ≥5 after 72 hours following parturient approval and consent, or if requested by the parturient at any time. Subsequent management after the study period, including EBP and adverse effects, were recorded. The primary outcome was the VAS at 24 hours. Additional predetermined outcomes were the requirement for an EBP, neck stiffness, nausea and vomiting, and any adverse effects associated with the neostigmine/atropine mixture.

Statistical Analysis

The aim of this study was to determine the differences in the VAS primary outcome and the secondary outcomes including the need for an EBP, neck stiffness, and nausea and vomiting in the neostigmine/atropine and control groups. Results were expressed as means ± SD, medians with interquartile range, or numbers and percentages of participants as appropriate. The Hodges–Lehmann estimate was used to calculate the median difference for VAS values between the experimental and placebo groups. The primary outcome (VAS) in the 2 study groups was compared by linear mixed-effects (between-within group) repeated measures model with adjustment of baseline as covariate to assess differences over time and the group-by-time interaction.

Categorical data (neck stiffness and nausea and vomiting) were evaluated by χ 2 or Fisher exact test when appropriate. A binary logistic regression was performed to compare neck stiffness and nausea and vomiting between groups at 72 hours with adjustment for baseline variable at 0 hour as covariate. We selected the time point 72 hours as a priority point being the point at which blood patch is deemed necessary to manage PDPH after failure of the medical treatment in the study groups.

Patient age, weight, height, body mass index, and onset of headache were compared by the independent Student t test. The Shapiro–Wilk test was used to verify the normality of continuous data distributions. P values <.05 were considered significant. Statistical analysis was performed using the statistical package for the social sciences version 22 (SPSS Inc, Chicago, IL).

A pilot study was performed prior to patient recruitment to estimate an appropriate sample size. The pilot study included 20 subjects, 10 in each arm. The smallest effect size d that would lead to a clinically significant difference was found to be equal to 0.63. Effect size was derived by software after entering mean difference and SD to calculate the sample size. An estimated mean VAS of 7.15 in the placebo group and 5.64 in the neostigmine group, with a pooled SD of 2.41, were used in the sample sized calculation. A sample size of 34 participants provided a 2-tailed α = .05, 80% power (â = .2), and an allocation ratio = 1. Forty-five participants were included to account for possible protocol violations or loss of data. The sample size calculation was performed with G*Power software version (Institute of Experimental Psychology, Heinrich Heine University, Dusseldorf, Germany).[19]