The pattern of blood loss during correction surgery combined with three column osteotomy for adult spine deformity

Background: It is of great importance to perform a precise management for blood loss according to stage of surgery. To analyze blood loss during different stages of three column osteotomy (3CO) surgery for scoliosis and to find predictors of blood loss during different stages and to evaluate effectiveness of some blood conservative technique modalities. Methods: A retrospective review was performed for adult spine deformity (ASD) patients who underwent posterior-only correction along with 3CO. Surgical procedure was divided into four stages: Stage A, exposure; Stage B, screw placement; Stage C, osteotomy; Stage D, rod installment and wound closure. EBL was recorded at the end of each stage. Demographic and radiographic data were collected. Results: Highest EBL was observed at neuromuscular scoliosis group, while least at idiopathic. 9 patients experienced massive blood loss (EBL>5000 ml). Multivariate linear regression identified fusion level main curve Cobb angle and use of TXA as influential factors for stage A; fusion level, use of TXA and use of navigation system for stage B; kyphosis and use of bone scalpel for stage C; fusion level and main curve Cobb angle for stage D (P<0.05). Conclusion: Etiology significantly influences blood loss of scoliosis surgery with 3CO. Risk factors of blood loss were varied among different surgical stages. Use of TXA, navigation system and ultrasonic bone scalpel could reduce blood loss.


Introduction
For rigid scoliosis, it is impossible to be corrected either with instrumentation alone or with the addition of posterior ligament and simple facet joint releases 1 . Vertebral osteotomies allow the surgeon to obtain more correction in rigid deformities 2 . There are three main types of osteotomies, including the Smith-Petersen osteotomy (SPO), pedicle subtraction osteotomy (PSO), and vertebral column resection (VCR) 3 . PSO and VCR are often termed as three-column osteotomies (3CO) and considered as procedures with high risks of complications 4,5 . 3CO was often associated with massive blood loss, sometimes above 10L 6 . Massive intraoperative blood loss increases rate of infections, hemodynamic instability, decreased cardiac, pulmonary, and renal function secondary to fluid shifts, and even death 7 . During procedures of 3CO, massive blood loss causes hypotension, and then increases the risk of neurologic deficits 8 . Some blood conservative technique modalities including the use of ultrasonic bone scalpel, autologous blood donation, controlled hypotensive anesthesia, preoperative administration of erythropoietin, intraoperative blood salvage, intrathecal morphine and administration of specific agents like tranexamic acid during the operation have been adopted to reduce blood loss and blood transfusion [9][10][11] . However, these modalities are costly and unavailable at some regions. Unlike other scoliosis surgery, blood loss during 3CO was not linear, and the pattern of blood loss was varied during different stages of surgery. It is of great importance to perform a precise management for blood loss according to stage of surgery. The objectives of this study were:1) to analyze blood loss during different stages of 3CO surgery for scoliosis; 2) to find predictors of blood loss during different stages; 3) to evaluate effectiveness of some blood conservative technique modalities. Total intravenous anesthesia was administered to all patients. Blood pressure and heart rate were controlled to fluctuate within 20% of the basic level. The intraoperative body temperature was kept between 36℃ and 37℃ with the use of warm air blower.

Materials and Methods
Cell salvage autologous blood recovery system was used in some cases. Tranexamic acid (TXA) was used to reduce postoperative blood loss. The standard TXA dose used in our study was 100 mg/kg intravenously over 15 min after induction of anesthesia and before incision followed by an infusion of 10 mg/kg per hour during surgery until skin closure 12 .
Transfusion decisions were made in the following conditions: 1) symptomatic anemia such as hypotension that was inadequately responsive to fluid challenge; 2) hemoglobin concentration of less than 80 g/L; or 3) hemoglobin level less than 100 g/L with signs of hemodynamic instability (e.g. tachycardia, poor peripheral perfusion).
Estimated blood loss (EBL) was determined by calculating the blood volume in the suction container after subtracting irrigation in addition to the difference in the weights of dry and blood-soaked sponges. Surgical procedure was divided into four stages: Stage A, exposure; Stage B, screw placement; Stage C, osteotomy; Stage D, rod installment and wound closure. EBL was recorded at the end of each stage. Time of each stage was also recorded. Demographic data such as age at the time of surgery, sex and body mass index (BMI) were collected. Radiographic parameters included main curve Cobb angle, kyphosis, osteotomy level and fusion levels. Other factors influencing blood loss were also recorded, such as use of navigation system, use of ultrasonic bone scalpel and use of TXA.

Statistical analysis
The correlation between the measured variables and EBL at each stage was determined using Pearson's correlation coefficient (r) and Spearman's rank correlation coefficient (rho). Variables with a significant correlation were then analyzed using a stepwise multivariate linear regression. A P value of 0.05 was set to indicate statistical significance. Massive blood loss was defined as EBL larger than 5000ml. A total of 9 patients experienced massive blood loss. Of the 9 patients, 3 cases were neuromuscular, 3 degenerative, 2 congenital, 1 syndromic. Table 2

Discussion
Corrective surgery for scoliosis puts the patient at a risk of massive blood loss 13 .
Excessive intraoperative and postoperative blood loss increases the transfusion burden on the patient, causing increased exposure to blood products and its associated risks 14,15 . It is of great importance to identify risk factors and adopt the necessary measures to optimize blood management. Numerous studies have been performed and identified many risk factors of blood loss for scoliosis corrective surgery. However, these studies did not take account of surgery stage as influencing factor for blood loss. In fact, the pattern of blood loss in corrective surgery was varied between different surgical stages. 3CO may be the biggest source of blood loss for scoliosis corrective surgery, even above 10L in some cases.
We divided the whole surgery to four stages, and identified risk factors for each stage, facilitating performing more meticulous blood management.
Etiology significantly influences blood loss of scoliosis surgery 13,16 . Operative treatment of AIS is typically associated with lower amounts of mean blood loss than surgical correction of scoliosis secondary to an underlying neuromuscular disorder 17 . Patients with neuromuscular scoliosis may be predisposed to higher intraoperative blood loss by various other factors, including poor nutritional status, treatment with antiepileptic medications, altered connective tissue hemostatic and coagulation properties, qualitative platelet disorder, decreased venous tone, and osteopenic bone 18 . In our study, idiopathic scoliosis had less blood loss as compared to other types of scoliosis. Idiopathic scoliosis treated by 3CO always presented as a round kyphosis, and osteotomy site closure was easier. Complete closure of osteotomy and reduction of operative time can significantly reduce the amount of bleeding during 3CO.
We divided the whole surgical procedure to four stages: exposure, screw placement, osteotomy and rod installment and wound closure. At exposure stage, fusion level, main curve Cobb angle and use of TXA were influential factors. It is not surprised that long severe scoliosis would cost more time for exposure, which would increase blood loss. The efficacy of TXA on reducing blood loss has been widely accepted 11,12 . TXA take effects at the beginning of surgery, significantly reducing blood loss of exposure. At screw placement stage, two factors determine the blood loss: number of placed screws and complexity of screw placement. Number of screws was positively correlated to fusion level, which was identified as an influential factor in our study. Extremely thin and deformed pedicles were frequently seen at neuromuscular and congenital scoliosis, which pose great challenges to surgeons. Surgeons often try several times when placing pedicle screws at these pedicles, and prolonged time for placement would significantly increase blood loss. In addition, intravertebral vein plexus were always damaged when pedicles were violated during pedicle placement, causing violent bleeding. Oarm based navigation system could greatly improve accuracy of screw placement deformed pedicles 19 . In our study, use of navigation system was negatively related to blood loss. We recommend using navigation system in difficult screw placement which would not only boost safety but also reduce blood loss. Highest blood loss was observed at osteotomy stage. The main source of bleeding was interverbal venous plexus and bone surface of osteotomized vertebra 20 . Our study demonstrated that kyphosis and use of bone scalpel were related to blood loss.
More sever was kyphosis, more time was needed for osteotomy. Before closure of osteotomy site completed, bleeding from interverbal venous plexus and bone surface continuously occurred. Ultrasonic bone scalpel could cut out laminae in a regular shape limiting the harassment to intervertebral venous plexus. Moreover, the friction between the bone scalpel blade and the bone is believed to reduce blood loss by sealing the cut bone through the heat emitted from the mechanical energy created by the rapid, repetitive movement of the scalpel 9,21 . At the last stage, blood loss was determined by the completion of bone closure and length of incision. Bone closure in patients with large coronal curve was difficult, and blood would be continuously lost from incompletely closed bone surface.
There was no a well-accepted definition of massive blood loss. We defined massive blood loss as EBL more than 5000ml. Some surgeons advocated defining massive blood loss as EBL more than 30% of the estimated blood volume [22][23][24] . However, this definition was not suitable for our cohort, as most cases had EBLs from 1000 to 2000ml. Majority of patients in our cohort with abnormal blood loss had EBLs more than 5000 ml. No idiopathic scoliosis patient experienced massive blood loss.
Highest blood loss was observed in a patient with degenerative scoliosis, who had an EBL of 10100ml. The reason why so high blood loss occurred was that surgeon injured lumbar artery during osteotomy. The fierce bleeding was controlled by embolizing injured artery under DSA. For neuromuscular patients with massive blood loss, fierce bleeding started from stage A, while mainly at stage C and D for degenerative scoliosis. As stated, patients with neuromuscular scoliosis may be predisposed to higher intraoperative blood loss by various other factors, including poor nutritional status, treatment with antiepileptic medications, altered connective tissue hemostatic and coagulation properties, qualitative platelet disorder, decreased venous tone. For patients with degenerative scoliosis, higher blood loss was always caused by bleeding from bone surface due to osteoporosis and altered connective tissue hemostatic.