Study design
This was a double-blind randomized placebo-controlled study on the efficacy and safety of intravenous TXA at reducing blood loss in women undergoing vaginal delivery at a tertiary centre southeastern Nigeria who met the inclusion criteria after obtaining an informed consent. The study lasted between June 2019-December 2019. The study was approved by was approved by the Human Research and Ethics Committee (HREC) of Alex-Ekwueme Federal Teaching Hospital, Abakaliki (FETHA/REC/VOL1/2017/541) and registered with Pan African Clinical Trial Registry: (PACTR202010828881019).
Study setting
The Federal Teaching Hospital Abakaliki (FETHA) was established in December 2011 following the merger between the defunct Federal Medical Centre and the then Ebonyi State University Teaching Hospital. FETHA has eleven clinical departments including Obstetrics and Gynaecology. The department of Obstetrics and Gynaecology of the hospital runs antenatal clinics managed by consultants and resident doctors with trained Nurses and Midwives. Antenatal clients are booked daily; Monday through Friday and patients are assigned to consultants according to the units/teams running antenatal clinic each day. The department has five units (each divided into two teams). Each team is manned by at least two consultant staff. The department manages both low and high risk pregnant women using standardized protocols.
Participants
The participants for this study were from the population of women within the reproductive age group admitted to undergo vaginal delivery at the Alex-Ekwueme Federal University Teaching Hospital, Abakaliki who met the inclusion criteria after obtaining an informed consent.
The inclusion criteria include spontaneous labour in booked patients, planned vaginal delivery, term pregnancy, singleton pregnancy and cephalic presentation, parturient who has no contraindication to the use of tranexamic acid and informed consent form signed.
Women with prior history of thromboembolism/autoimmune/ sickle cell disease, bleeding disorders, renal disease, liver pathology, known cardiovascular disease, multiple pregnancy/intra-utero fetal death/previous uterine surgeries, patients with chronic hypertension, preeclampsia/eclampsia/HELLP syndrome, antepartum haemorrhage, ruptured uterus, varicose veins at increased risk of deep vein thrombosis, history of epilepsy/seizure and those that had episiotomy were excluded.
Sample size
The minimum sample size was determined using the formula for comparison between two groups when the end point is a quantitative data29
Sample size = 2SD2 (Zα/2 + Zβ) 2
d 2
Where:
SD: standard deviation in blood loss from treatment group 3928= 0.39
Zα/2 : Standard normal deviate at 5% type 1 error= 1.96
Zβ : To increase accuracy of the study 90% power was used = 1.282
d: Standardized effect size 21% reduction in blood loss28= 0.21
Sample size per group = 2 (0.39)2 x (1.96 + 1.282)2 /(0.21)2
Twenty percent (20%) of the minimum sample size per group (20/100×73/1≈15) was added to correct for any attrition hence the final sample size was 88 for each arm.
Randomization And Concealment
The participants were randomized by means of a computer-generated random number using the software Research Randomizer®. Eighty-eight (88) numbers were randomly generated from a pool of one hundred and seventy-six (1-176) and these numbers were assigned to group A (tranexamic acid group), while the remaining eighty-eight were automatically assigned to group B (the placebo group).
Group A received 1g tranexamic acid (Exacyl®; Sanofi Aventis Paris France) slowly (over 30-60 seconds) intravenously, within 2 minutes after birth and prophylactic oxytocin administration once the cord had been clamped.
Group B received 10mls of water for injection (Biofem®; Juhel Anambra Nigeria) slowly (over 30-60 seconds), within 2 minutes after birth and prophylactic oxytocin administration, once the cord had been clamped. These drugs were sourced from their drug representatives.
Concealment was done in sequentially numbered opaque sealed envelopes (SNOSE).32 These numbers (1-176) were inscribed on brown envelopes and a piece of paper with the inscription ‘tranexamic acid’ or ‘placebo’ was placed with the respective drug or placebo accordingly inside these envelopes and sealed. The randomization was done by a statistician and me, while the concealment was done by a hospital pharmacist without revealing the results to the researcher. All the envelopes were kept in a locker that was made accessible to all the members of the research team.
Participants that met the inclusion criteria having signed the informed consent form were given sequential study number and the corresponding numbered opaque sealed envelope was allocated to the patient.
Study Procedure
Women were selected for vaginal delivery in the facility and admitted into the labour ward. They were counseled on the study and those who signed the informed consent form were recruited. The antenatal card was retrieved, and highlights reviewed. History was taken and clinical examination was done to confirm the stage of labour while ancillary investigations; haematocrit, haemoglobin and urinalysis were done and patients were transferred into the labour ward in active phase of labour. The labour was managed actively with the partograph, and augmentation was done as indicated.
The researcher or any of the research assistants took the allotted sealed envelope to the labour ward and handed same over to the labour ward officer who administered the drug or the placebo over 30 to 60 seconds within 2 minutes of delivery of the baby. The envelope with its used content (resealed) was returned to the investigators who kept all the used envelope/packs in a separate locker until the end of the study when un-blinding was done.
AMTSL was carried out for all recruited patients according to departmental protocol (cord clamping, use of oxytocin and controlled cord traction).28 Other oxytocics and surgical interventions required to control excessive bleeding were given or done and patients who needed blood transfusion received same.
These interventions were noted. Immediately after the delivery of the baby, when all the liquor were drained, a blood drape (an improvised BRASS-V, a disposable conical, graduated plastic collection bag) was inserted under the patient.28,32 This was locally manufactured. The blood collected in the blood drape was transferred into a transparent plastic measuring cylinder with a capacity of 500ml, corrected to 2ml and manufactured by Measure Masters®. The blood in the measuring cylinder was read off and documented by the researcher or the assistants. Then, the patient was given pre-weighed pads, which were re-weighed 2 hours post-partum.28 For uniformity, the regular labour ward sanitary pad with negligible dry weight was used. EBSA-20 electronic weighing scale, which operates at room temperature with readability of 20kg/5g for maximum and minimum weights respectively and manufactured by the Zhongshan Jimli Electronic Weighing Equipment Co. Ltd was used.
The side effects of the drug were noted. The patients were transferred to the post-natal ward for further observation. The patients’ post-delivery pulse rate and blood pressure were noted and recorded.32
Blood Loss Estimation
The estimated blood loss ascertained by measuring the blood collected in the drape and complimented by measuring the weights of the sanitary pads before and after 2hrs of delivery. Immediate post-partum blood loss was calculated thus.32
Total blood loss (ml) = Blood in the measuring cylinder (ml) +[ Pad weight after 2 hours (gm)-Pad weight prior to use (gm)]- converted to ml; taking that 1 gram difference in pad weight equals 1 ml of blood.33
Blood loss greater than >500ml was be regarded as excessive bleeding.Investigating Maternal Mortality in a Public Teaching Hospital, Abakaliki, Ebonyi State, Nigeria
Follow up
Patients were expected to stay for 48 hours on admission in the hospital before discharge except otherwise indicated. The duration of stay was dependent on the patient’s clinical state. The participants were followed up until discharge from the facility. They were instructed to present to the hospital or reach the researcher or any of the research assistants by phone if they have any unforeseen adverse reaction which was to be reported.
Outcome measures
Primary outcome measure was estimated blood loss following vaginal delivery (total blood loss following vaginal delivery= estimated blood from cylinder+ difference in the weight of pad)32. Secondary Outcome Measures include primary PPH following vaginal delivery defined as blood loss > 500ml, need for additional uterotonics to control bleeding, need for blood transfusion (volume and amount) after vaginal delivery, mild maternal side effects (nausea, vomiting, headache, skin rash), major maternal side effects (thromboembolism, maternal death).31
Statistical analysis
Data was collated, tabulated, and statistically analyzed with the Statistical Package for Social Science (IBM SPSS) software (version 20, Chicago II, USA). Continuous variables as the maternal vital signs were presented as means and standard deviations (Mean ± 2SD), while categorical variables like minor and major side effects were presented as numbers and percentages. Chi-square test (X2) was used for comparison between groups for qualitative variables while T-test was used for comparison between groups for quantitative variables. A difference with a p-value <0.05 was considered statistically significant.