Introduction

Cell Saver is commonly used in spinal fusion for scoliosis as a means of salvaging and returning the patient’s red blood cells. However, it is not clear whether use of Cell Saver decreases the need for other transfusions in this patient population. While some studies have found that Cell Saver use decreases the need for other blood transfusion in orthopedic surgery [1–4], others have found its use to be of little or no benefit [5, 6].

The device itself, setup, and personnel associated with Cell Saver use add significant cost to the surgery. Others have estimated the cost of Cell Saver for spinal fusion to be $240 [6] to $512 [5] per operative case. These authors concluded that Cell Saver was more expensive than predonated blood for autotransfusion [5, 6].

It is unclear whether there are certain patients who would most likely benefit from the use of Cell Saver. Previous studies examining the usefulness of Cell Saver in orthopedic surgery have included mostly adult patients [1–5], with the exception of Siller et al. [6], who studied adolescents undergoing spinal fusion for idiopathic scoliosis. To our knowledge, there has been no published study examining scoliosis patients of all ages and etiologies to determine the usefulness of Cell Saver in this setting.

Furthermore, red blood cells collected by the Cell Saver system are not always of sufficient volume to be returned to the patient. At our institution, 250 cc of blood must be collected to facilitate the wash necessary for return. It is unclear how much blood is actually lost for 250 cc to be collected, washed, and returned to the patient.

The purpose of this study was to determine whether (1) Cell Saver reduces the need for other (non-Cell Saver) transfusions in posterior spinal fusion for scoliosis, (2) there are certain patients in whom Cell Saver blood is more likely to be given back intraoperatively, and (3) there is an amount of blood loss at which Cell Saver is likely to be returned.

Materials and methods

Two surgeons performed a total of 95 consecutive cases of posterior spinal fusion for scoliosis during a 2-year period from January 2002 to March 2004. Both surgeons operated in the same hospital under similar conditions. These surgeons worked closely together and employed similar techniques in exposure, hardware placement, and closure. They consulted with one another in pre-, intra-, and postoperative planning and decision making. Patients were excluded from the study if they underwent an anterior procedure or were opposed to blood transfusion due to religious beliefs (Jehovah’s witnesses). The patients were not randomized, and use of the Cell Saver was determined by the surgeons' preference. One surgeon used Cell Saver in every spinal fusion for scoliosis, the other did not. Blood loss was estimated by evaluating the amount of blood in the suction canister and that in the soaked lap pads. Patients with idiopathic scoliosis received a blood transfusion if they showed clinical signs or symptoms of hypotension and had a hematocrit of less than 20. Patients with neuromuscular scoliosis received a blood transfusion if they showed clinical signs or symptoms of hypotension or had a hematocrit of less than 25. Two drains were routinely placed postoperatively, one superficial and one deep. Drains were discontinued no sooner than postoperative day 3, with output of less than 100 cc over 24 h. The Cell Saver group included 58 patients. The control group, in which Cell Saver was not used, included 37 patients.

Patients’ charts were reviewed retrospectively for age, gender, diagnosis, weight, number of levels fused, intraoperative blood loss, postoperative blood loss, blood returned from the Cell Saver, and other blood transfusions. The term “other blood transfusions” is used to describe all transfusions that were not Cell Saver return.

The data were analyzed to determine which factors were associated with higher rates of Cell Saver return.

Results

The average patient age was 14 years plus 1 month, with patients of similar ages in each group. There were 25 male patients and 70 female patients, with similar distributions in each group. The patients’ average weight was 50.3 kg, with no significant difference between the groups (P = 0.365). Demographics are further detailed in Table 2.

The mean pre-operative Cobb angle was 63.9° in group 1 and 63.1° in group 2 (Table 1). The average number of levels fused was 11.8 levels. The average in the Cell Saver group was 11.2 levels, and the average in the control group was 12.7 levels (P = 0.058) (Table 1).

Table 1 Age, estimated blood loss, weight, and number of levels fused
Full size table

Intraoperative blood loss averaged 733 cc. There was no statistical significance in blood loss between group 1 (EBL 763 cc) and group 2 (EBL 688 cc) (Table 1). There was no correlation between number of levels fused and the likelihood of receiving IAT (P = 0.18) or between number of levels fused and likelihood of receiving other transfusions.

Patients in the Cell Saver group were just as likely to receive non-Cell Saver transfusions as those in the control group (P = 0.12). Of the 39 patients who lost more than 500 cc of blood, 34 received IAT (87%). No patient who lost less than 500 cc received IAT.

The diagnoses were as follows: 71 patients had idiopathic scoliosis and 24 patients had neuromuscular scoliosis. Specific neuromuscular diagnoses included 15 patients with cerebral palsy, 6 patients with muscular dystrophy, 2 patients with Rett’s syndrome, and 1 patient with neurofibromatosis (Table 2). Patient distribution between groups 1 and 2 was even. There was no correlation between diagnosis and non-Cell Saver transfusions. Children in group 1 with neuromuscular scoliosis were just as likely to receive non-Cell Saver transfusions as those in group 2 (P = 0.26). Children in group 1 with idiopathic scoliosis were just as likely to receive non-Cell Saver transfusions as those in group 2 (P = 0.16).

Table 2 Demographics: summary of patient data (missing data denoted by empty boxes)
Full size table

The mean pre-operative hemoglobin/hematocrit was 13.5/39.7 (group 1 mean 13.5/39.5, group 2 mean 13.4/39.8). The mean hemoglobin/hematocrit on postoperative day 1 was 9/27 (group 1 mean 9.2/27.7, group 2 mean 8.8/26.1). The mean hemoglobin/hematocrit upon discharge was 10/30 (group 1 mean 10/30, group 2 mean 10/30) Table 2.

Discussion

Cell Saver is widely used in scoliosis surgery. The benefit of the device has not been proven, however. It was our anecdotal experience that Cell Saver does not consistently decrease the patient’s need for other transfusions. This study was thus undertaken to examine the use of Cell Saver in scoliosis surgery.

One weakness of this investigation was that the decision to transfuse the patient was dependent on the attending surgeons' preference. This was based on the clinical picture, and was thus not standardized.

Previous studies have shown that Cell Saver does decrease the need for transfusions of other kinds [7–10]. Goulet et al. [2] found that transfusion requirements were decreased when Cell Saver was used in orthopedic procedures. McMurray et al. [1] and Garvin et al. [11] found that Cell Saver decreased the need for donor blood transfusion in primary and revision hip arthroplasty. Cell Saver has also been shown to decrease the amount of other blood transfusions in children undergoing acetabuloplasty [12]. A meta-analysis of the literature on the topic of anesthesia examining the effectiveness of Cell Saver found that its use in orthopedic surgery decreased the need for other transfusions [3]. Reitman et al. [5] studied Cell Saver use in adult lumbar fusion surgery. They found that 36% of patients who had Cell Saver required 1 unit of blood transfusion, compared with 50% of patients who did not have Cell Saver. However, the authors were careful to qualify that autologous blood predonation affected their study, and that “the use of Cell Saver in patients who did not predonate (blood) was not significantly beneficial toward reducing the need for transfusion”. Anand et al. [13] showed that the use of Cell Saver decreased the need for homologous blood transfusion.

We conclude that Cell Saver does not decrease the incidence of other blood transfusions in scoliosis surgery. Other studies that draw similar conclusions include the following: Siller et al. [6] examined Cell Saver use in spinal fusion for idiopathic scoliosis and found that “blood requirements for this procedure can be met less expensively and more reliably by merely donating one’s own blood”. Similar findings are reported regarding other orthopedic procedures and Cell Saver. Lisander et al. [4] concluded that Cell Saver is not sufficient to decrease the need for other blood transfusions in hip arthroplasty. Abildgaard et al. [14] came to a similar conclusion regarding the use of Cell Saver in major back surgery.

The Cell Saver device is often set up for surgery without return of the collected fluid to the patient. This study found that blood loss of 500 cc correlated with return of blood to the patient using this system. The surgeon can use this number to better estimate the likelihood of Cell Saver return when deciding whether or not to use the device. If less than 500 cc of blood loss is expected, this study shows that the cost of Cell Saver setup and personnel is not justified.

During this study, we found no correlation with benefit based on number of levels fused, diagnosis, or patient weight. These criteria are therefore not helpful in aiding the surgeon in planning for Cell Saver setup.

Conclusion

The use of Cell Saver does not reduce the need for other transfusions in scoliosis surgery. The amount of blood loss at which Cell Saver is likely to be returned is 500 cc. We demonstrated no benefit in the use of Cell Saver in our patient population.