Table 1 lists the cohorts’ baseline demographics (mean age, BMI, preoperative Hb level) and comparative ANOVA results. TOSTs of equivalence were performed to compare operative time, EBL, 48-hour wound drainage, and postoperative Hb-level depreciation among hemostasis types. Changes in Hb level were calculated for the immediate postoperative period and time of discharge (Table 2). ANOVA of hospital length of stay demonstrated no significant difference in means among groups (P = .09).
The cohorts were compared with respect to use of postoperative transfusions and incidence of postoperative AEs (Table 3). The TE cohort did not have any AEs. Of the 203 BS patients, 14 (7%) had 1 or more AEs, which included acute kidney injury (3 cases), electrolyte disturbance (3), urinary tract infection (2), oxygen desaturation (2), altered mental status (1), pneumonia (1), arrhythmia (1), congestive heart failure exacerbation (1), dehiscence (1), pulmonary embolism (2), and hypotension (1). Of the 36 ABC patients, 1 (3%) had arrhythmia, pneumonia, sepsis, and altered mental status.
Discussion
With the population aging, the demand for TKA is greater than ever.1 As surgical volume increases, the ability to minimize the rates of intraoperative bleeding, postoperative anemia, and transfusion is becoming increasingly important to patients and the healthcare system. There is no consensus as to which cautery method is ideal. Other investigators have identified differences in clinical outcomes between cautery systems, but reported results are largely conflicting.10,12-20 In addition, no one has studied the utility of ABC in TKA. In the present retrospective cohort analysis, we hypothesized that TE, BS, and ABC would be clinically equivalent in primary TKA with respect to EBL, 48-hour wound drainage, operative time, and change from preoperative Hb level.
The data on hemostatic technology in primary TKA are inconclusive. In an age- and sex-matched study comparing TE and BS in primary TKA, BS used with shed blood autotransfusion reduced homologous blood transfusions by a factor of 5.16 In addition, BS patients lost significantly less total visible blood (intraoperative EBL, postoperative drain output), and their magnitude of postoperative Hb-level depreciations at time of discharge was significantly lower. In a multicenter, prospective randomized trial comparing TE with BS, adjusted blood loss and need for autologous blood transfusions were lower in BS patients,10 though there was no significant difference in Knee Society Scale scores between the 2 treatment arms. However, analysis was potentially biased in that multiple authors had financial ties to Salient Surgical Technologies, the manufacturer of the BS device used in the study. Other prospective randomized trials of patients who had primary TKA with either TE or BS did not find any significant difference in postoperative Hb level, postoperative drainage, or transfusion requirements.19 ABC has been studied in the context of orthopedics but not joint arthroplasty specifically. This technology was anecdotally identified as a means of attaining hemostasis in foot and ankle surgery after failure of TE and other conventional means.22 ABC has also been identified as a successful adjuvant to curettage in the treatment of aneurysmal bone cysts.21 However, ABC has not been compared with TE or BS in the orthopedic literature.
In the present study, analysis of preoperative variables revealed a statistically but not clinically significant difference in BMI among cohorts. Mean (SD) BMI was 35.6 (6.5) for TE patients, 35.8 (9.7) for BS patients, and 40.9 (11.3) for ABC patients. (Previously, BMI did not correlate with intraoperative blood loss in TKA.25) Analysis also revealed a statistically significant but clinically insignificant and inconsequential difference in Hb level among cohorts. Mean (SD) preoperative Hb level was 13.5 (1.6) g/dL for TE patients, 12.8 (1.4) g/dL for BS patients, and 13.0 (1.6) g/dL for ABC patients. As decreases from preoperative baseline Hb levels were the intended focus of analysis—not absolute Hb levels—this finding does not refute postoperative analyses.
Our results suggest that, though TE may have relatively longer operative times in primary TKA, it is clinically equivalent to BS and ABC with respect to EBL and postoperative change in Hb levels. In addition, postoperative drainage was lower in TE than in BS and ABC, which were equivalent. No significant differences were found among hemostasis types with respect to postoperative transfusion requirements.
The prevalence distribution of predischarge AEs trended toward significance (χ2 = 5.957, P = .051), despite not meeting the predetermined α level. Rates of predischarge AEs were 0% (0/41) for TE patients, 7% (14/203) for BS patients, and 3% (1/36) for ABC patients. AEs included acute kidney injuries, electrolyte disturbances, urinary tract infections, oxygen desaturation, altered mental status, sepsis/infections, arrhythmias, congestive heart failure exacerbation, dehiscence, pulmonary embolism, and hypotension. Clearly, many of these AEs are not attributable to the hemostasis system used.