Effect of Tranexamic Acid on Blood Loss and the Quality of Surgical Field in Transsphenoidal Pituitary Surgeries: Double-Blind Placebo-Controlled Randomized Control Trial
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.349622
Source of Support: None, Conflict of Interest: None
Keywords: Boezaart score, pituitary surgery, quality of surgical field, tranexamic acid, transsphenoidal surgery
Pituitary adenomas are common, diverse group of tumors of the pituitary gland. Transsphenoidal route is the most common neurosurgical approach of resection of these adenomas. This route is preferred for all but the largest of tumors, as it has the advantage of rapid midline access to the sella with minimal risk of brain trauma., The goal of neuroanesthesia in these surgeries is to provide stable hemodynamics, a clear surgical field, adequate cerebral oxygenation, and rapid emergence to facilitate early neurological assessment.
Risk of bleeding is usually modest in transsphenoidal pituitary surgeries. However, even slight bleeding can cause considerable difficulties for the surgeon by obscuring the view of the surgical field through the operating microscope or endoscope. With increased bleeding, the neurosurgeon can lose the anatomic landmark of the fossa and might have difficulty differentiating normal tissue from tumor. This can lead to catastrophic complications, such as injury to the neighboring structures. These surgeries have the potential for massive intraoperative bleeding because of the proximity of the pituitary gland to the carotid arteries. Profound blood loss can also occur from continuous oozing from the cavernous sinus and nearby capillaries.
There are very few studies that have focused on minimizing blood loss in endoscopic or microscopic intracranial procedures. The use of antifibrinolytics such as tranexamic acid to reduce bleeding and transfusion requirement is widespread in cardiac, pediatric, orthopedic, and spinal surgeries.,,, Tranexamic acid is a synthetic derivative of the amino acid lysine that competitively blocks the breakdown of fibrin by binding to sites on plasminogen and plasmin, thereby stabilizing blood clots and reducing blood loss. Even though TXA use has shown to reduce intraoperative blood loss in various surgeries, including skull base, cranial vault, sinus and open intracranial tumor surgeries, it has never been used in a randomized study for pituitary surgeries.
After the Institute's Ethics Committee approval, a randomized, double blind, placebo-controlled study was conducted in 50 patients undergoing elective transsphenoidal pituitary surgery. Patients aged 18-75 years, American Society of Anesthesiologists physical status I and II, of either sex were included in the study. Exclusion criteria included patients with pituitary apoplexy, impaired renal function (creatinine >1.5 mg/dl), preoperative hemoglobin less than 8 gm/dl, patients on anticoagulant or antiplatelet medications and those who had a previous pituitary surgery. Intraoperative blood transfusion was guided by our institutional practice like measured hemoglobin less than 8 gm/dl and sudden massive bleeding with hemodynamic instability. Written informed consent was obtained from all the patients before enrolment in the study.
Simple random allocation rule was followed for randomization. Patients were randomly recruited into either Group T or Group P by a computer-generated random number table. The random numbers were then kept in opaque, sealed envelopes and numbered sequentially. The anesthesiologist who prepared the study drug did not participate in anesthetizing the patient or collecting the data. The anesthesiologist managing the patient and collecting the data along with the operating neurosurgeon were blinded to the study drug. A single blinded neurosurgeon performed all the surgeries to ensure consistency in estimating the quality of the surgical field.
An identical standard anesthesia protocol was followed for all patients. All patients were kept nil per oral for 8 hours. Intravenous fentanyl 2 μg/kg, propofol 1-2 mg/kg and vecuronium 0.1 mg/kg were used at induction of anesthesia. Anesthesia was maintained by a titrated infusion of propofol and intermittent doses of vecuronium and fentanyl as required. The patients were ventilated to maintain end-tidal CO2 at 35 to 38 mmHg. Positive end-expiratory pressure (PEEP) was not used in any patient. After tracheal intubation, an arterial line was established in the radial artery.
In order to correct the fasting fluid deficit for baseline hematocrit estimation, 10 ml/kg of normal saline was first administered to all the patients in 15-20 minutes. After 10 minutes of completion of fluid administration, a blood sample was sent to the laboratory for hematocrit estimation. This hematocrit value was considered as “Initial hematocrit (Hi)”. Patients in group T then received a bolus dose of tranexamic acid 25 mg/kg over 10 minutes before the start of surgery, and then a continuous infusion of 1 mg/kg/hour through an infusion pump till the end of the surgery. Patients in group P received a corresponding volume of normal saline in an identical fashion. At all times, mean arterial pressure was maintained within 20% of the baseline value. The infusion of tranexamic acid was stopped at the end of the surgery. The second blood sample was then sent for “Final hematocrit (Hf)” estimation. At the end of the surgery, the neurosurgeon rated the quality of the surgical field according to the Boezaart Scale [Table 1]. Blood loss was calculated using the Modified Gross Formula: [predicted blood loss = blood volume x (Hi - Hf)]. Blood volume was estimated as the product of body weight and 0.75 (or 0.70 for females).
Prior to conducting this study, a pilot study was carried out in 10 patients undergoing transsphenoidal surgery of pituitary tumors. The same anesthesia protocol was followed in all these cases. None of the patients were given the test drug. The mean intraoperative blood loss was found to be 470 ± 137 ml. In order to reduce the mean intraoperative blood loss satisfactorily by a factor of 30% using the test drug with an alpha error of 0.05 and a power of 80%, a sample size of 22 patients in each group was required. Considering the possibility of unforeseen circumstances, we chose a sample size of 25 in each group with a total sample size of 50 patients.
Statistical analysis was carried out using IBM SPSS, version 21 for Windows. Variables such as age, weight, total duration of surgery and anesthesia, total intravenous fluids administered, and requirement of anesthetic agents were compared using the t-test. Categorical variables such as sex, ASA physical status, number of patients with hypertension or diabetes, and the type of pituitary pathology were compared among the two groups using Chi Square test. The difference in the quality of the surgical field (Boezaart score) between groups was analyzed using the Mann Whitney U test. Total blood loss difference between groups was analyzed using the independent sample t-test.
A total of 63 patients were assessed for eligibility [Figure 1], out of which 13 patients were not enrolled because of meeting exclusion criteria, leaving 50 patients who were randomized into two groups of 25 each. The patients in the two groups were comparable in terms of age, body weight, gender, ASA physical status, total duration of surgery, total duration of anesthesia, and administered intravenous fluids [Table 2]. Other baseline characteristics such as the type of pituitary adenoma (broadly classified as functional and non-functional) and the presence of comorbid conditions like diabetes mellitus and hypertension were also similar in the two groups [Table 3].
Most patients who had a good surgical field (corresponding to a lower Boezaart score of 2 and 3) were in Group T. Six patients in Group P had a Boezaart score of 2, compared to 11 patients in Group T. On the other hand, 8 patients who received placebo had a high score of 4 (indicating moderate bleeding and a poor surgical field), compared to only 2 patients who received TXA [Figure 2]. The median (interquartile range) Boezaart score in Group T was 3 (1.0), compared to 3 (1.5) in Group P. This difference was statistically significant (P = 0.03) [Table 4]. None of our patients required intraoperative blood transfusion. The mean blood loss in Group T was 334 ± 101 ml, whereas it was 495 ± 226 ml Group P. This difference of approximately 32% was also statistically significant (P = 0.002, [Table 4]).
Numerous surgical complications have been reported during transsphenoidal resection of pituitary adenomas. Some of these complications such as injury to the internal carotid artery can have immediate life-threatening consequences. Even modest bleeding can significantly worsen the surgical field under the magnification of the operating microscope or endoscope, potentially leading to serious complications. Furthermore, minimizing intraoperative bleeding also reduces the need for blood transfusion, even though blood transfusion is uncommon in transsphenoidal resection of pituitary adenomas. There are serious concerns of blood transfusion such as transmission of infectious diseases, increased postoperative infection, dyselectrolytemia, allergic reactions, volume overload as well as immune modulation effects., Thus, it is important for the neuroanesthesiologist to maintain hemodynamic stability, adopt strategies to reduce intraoperative bleeding, and provide a clear surgical field to facilitate complete and uncomplicated resection of the pituitary adenoma.
Antifibrinolytic agents, which inhibit the enzymatic breakdown of fibrin, are commonly used to reduce intraoperative bleeding. Fibrinolysis is an important physiologic process to maintain the patency of blood vessels and coexists with the coagulation system through an intricate network of feedback loops in the body. During trauma and surgery, where there is tissue destruction, this equilibrium shifts to promote fibrinolysis and thus, antifibrinolytic drugs are believed to exert their protective effect by decreasing bleeding. However, published literature on the use of TXA in neurosurgeries is scarce. Two recently published studies showed a significant decrease in blood loss with the use of tranexamic acid in patients undergoing craniotomy for tumor excision.,
Several methods and formulas have been designed to estimate blood loss, but accurate estimation isn't easy in the operating room setup. Estimation by visual inspection of the suction canister, wet sponges, pads, drapes and sheets is widely practised, but has a wide inter-observer variation and can often be very misleading. Many quantitative methods such as photo spectrometry, radioisotope dilution, platelet function analysis and accurate gravimetric methods are generally expensive, require special equipment, are time-consuming, and are either impractical or unavailable in most operating room setups.,,,, The technique used in our study suggested by Gross et al. is easy to perform and doesn't consume much time or expense. This cost-effective method has been used in several studies to assess blood loss.,,,,, In our study, the mean blood loss in Group T was less compared to Group P (P = 0.002). Although there was no need for transfusion in both groups, there was a significant absolute reduction of blood loss with TXA use, amounting to 32%.
Assessment of the view of the surgical field is subjective and mostly based on the qualitative assessment of the anesthesiologists or the surgeons. Various rating scales have been devised to assess the quality of the surgical field. The most commonly used is the one designed by Boezaart et al.,,, This is a pre-determined, objective category scale with six grades from 0 to 5: 0 (no bleeding) to 5 (severe bleeding requiring constant suctioning). In our study, more patients in Group T had a lower Boezaart score (score: 3 or less) compared to Group P. Similarly, a poor surgical field (score: 4 or more) was present in more patients receiving placebo compared to TXA. Similar findings were obtained by Alimian and colleagues in endoscopic sinus surgeries. In our study, the median Boezaart score was statistically better in those receiving TXA. This finding may be very useful from neurosurgical point of view.
There are a few limitations of this study. Dose-response study for tranexamic acid was not done. Similarly, the bolus and infusion doses of TXA were based on the review of similar studies carried out previously. However, these may not be the optimal doses in neurosurgical patients because large-scale studies on tranexamic acid in intracranial surgeries are lacking. Though there were no immediate complications seen in any of the patient in the study group, long-term potential adverse effects of the drug such as seizures and thromboembolic events, albeit rare, were not assessed. We did not use any other laboratory method of estimating blood loss due to logistical reasons. All the patients in the study had a normal platelet count and coagulation profile, and it is not known whether TXA remains equally effective in those with deranged coagulation profile or platelet abnormalities. Stratification of patients on the basis of tumor-size was not possible because of heterogeneity in the modality and reporting of the radiological scans, as well as significant time-lag between the scan and the surgery in some cases. Lastly, this is a single-center randomized study and larger multicentric studies are warranted for broad generalizations of these study findings.
This study not only shows a significantly better surgical field with the use of tranexamic acid but also a significantly lesser blood loss. In the doses administered, intraoperative complications were not encountered. Despite widespread use in other surgical specialties, tranexamic acid is not commonly used in pituitary surgeries. Routine use of this inexpensive drug can be considered in all transsphenoidal pituitary surgeries, unless a specific contraindication for its use exists.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]