Orthopedics

Blood Loss and Transfusion Rate in Noncemented and Cemented/Hybrid Total Hip Arthroplasty. Is There a Difference? A Comparison of 25 Matched Pairs

Michael E Trice, MD; Richard H Walker, MD; Beverly A Morris, RN, CNP; Clifford W Colwell, Jr, MD

Abstract

ABSTRACT

A common assumption among orthopedic surgeons is that cemented/hybrid total hip arthroplasty (THA) results in lower transfusion requirements than noncemented THA. The hypothesized mechanism to account for transfusion differences is that cement application, after polymerization, decreases perioperative blood loss. To evaluate this theory, a retrospective matchedpair study was performed to quantify perioperative, intraoperative, postoperative blood loss, and transfusion requirements among patients undergoing cemented/hybrid and noncemented THA. Twenty-five THA procedures were either cemented or hybrid and 25 were noncemented. Patients were matched with respect to age, gender, weight, and diagnosis.

Six patients in the noncemented group and 1 0 in the cemented/hybrid group received autologous blood in the Post Anesthesia Care Unit. Among this subgroup, those in the cemented/hybrid group received a significantly greater volume of transfused blood (519.20 ml, versus 291 .67 mL in the noncemented group, P<.017).

After stratifying patients by gender, additional analysis revealed a greater total blood loss in the male population than in females (1848 mL versus 1464 mL, P < .004). Males also had a greater drain volume (548 mL versus 414 mL, P < .02). The female population utilizing hormone replacement therapy (HRT) was found to have a lesser blood loss (810 mL versus 1209 mL, P < .0107) and received a smaller amount of transfused blood (1 1 64 mL versus 1 377 mL) than those who were not on HRT. Other significant findings included a shorter length of stay in the cemented/hybrid group (6.85 versus 7.56 days, P< .012).

Based on this matched-pair retrospective analysis of blood requirements, there is no indication that cemented/hybrid and noncemented THA procedures result in different blood requirements. However, differences were found in several areas when the population was stratified by gender. The information obtained from this study is important in its ability to solidify a foundation from which blood management decisions can be made.

Abstract

ABSTRACT

A common assumption among orthopedic surgeons is that cemented/hybrid total hip arthroplasty (THA) results in lower transfusion requirements than noncemented THA. The hypothesized mechanism to account for transfusion differences is that cement application, after polymerization, decreases perioperative blood loss. To evaluate this theory, a retrospective matchedpair study was performed to quantify perioperative, intraoperative, postoperative blood loss, and transfusion requirements among patients undergoing cemented/hybrid and noncemented THA. Twenty-five THA procedures were either cemented or hybrid and 25 were noncemented. Patients were matched with respect to age, gender, weight, and diagnosis.

Six patients in the noncemented group and 1 0 in the cemented/hybrid group received autologous blood in the Post Anesthesia Care Unit. Among this subgroup, those in the cemented/hybrid group received a significantly greater volume of transfused blood (519.20 ml, versus 291 .67 mL in the noncemented group, P<.017).

After stratifying patients by gender, additional analysis revealed a greater total blood loss in the male population than in females (1848 mL versus 1464 mL, P < .004). Males also had a greater drain volume (548 mL versus 414 mL, P < .02). The female population utilizing hormone replacement therapy (HRT) was found to have a lesser blood loss (810 mL versus 1209 mL, P < .0107) and received a smaller amount of transfused blood (1 1 64 mL versus 1 377 mL) than those who were not on HRT. Other significant findings included a shorter length of stay in the cemented/hybrid group (6.85 versus 7.56 days, P< .012).

Based on this matched-pair retrospective analysis of blood requirements, there is no indication that cemented/hybrid and noncemented THA procedures result in different blood requirements. However, differences were found in several areas when the population was stratified by gender. The information obtained from this study is important in its ability to solidify a foundation from which blood management decisions can be made.

Table

TABLE 1Postoperative outcomes

TABLE 1

Postoperative outcomes

ments are utilized in preparation for surgery. Patients are encouraged preoperatively to discontinue medications that could potentially increase rate of blood loss and/or suppress marrow production of blood cells. Erythropoietin, a pharmacologic modality, has also been used to stimulate red blood cell production. Another modality,10 preoperative acute normovolemic hemodilution (ANH), has been evaluated in the total joint arthroplasty population." ANH is as effective as preoperative autologous donation (PAD) combined with Cell Saver in minimizing transfusion requirements. Because ANH does not offer additional benefits to PAD plus Cell Saver, it is rarely used for elective orthopedic surgery.

Intraoperative measures have been perhaps the most important means of diminishing the incidence of allogeneic transfusion. Meticulous hemostasis has improved perioperative blood loss and decreased operative time.12 Cell Saver and other intraoperative autologous blood-replacement programs have proven themselves valuable in minimizing transfusion requirements.4,13,14 Hypotensive anesthesia is also an effective method for diminishing intraoperative blood loss.12,36

Minimizing postoperative blood loss has also been approached by several modalities without a definite consensus regarding the effectiveness of each method. For example, the use of postoperative drains has been rejected by some investigators who feel that drains increase overall blood loss.17 A second method, careful monitoring of postoperative medication regimens, ensures that iatrogenic hypocoagulable states do not increase transfusion risk.18 Lastly, reinfusion of blood lost postoperatively has been used with success.'4,17

Appropriate preoperative planning includes estimation of perioperative blood loss.19 Standard care in many parts of the country includes PAD for THA patients with a trend towards decreased donation volume. Early programs required donation of 3 to 4 autologous units,20 which has gradually decreased to a common standard of 2 units. More recently, surgeons have sought to decrease donation to 1 unit or even no units.

One point of contention in the orthopedic community is that blood loss in noncemented arthroplasty is greater than that in cemented/hybrid procedures. The theoretical basis for the "increased" blood loss is that noncemented procedures leave large bleeding bone surfaces while cemented/hybrid procedures cover diese surfaces with cement. This cement "caulking" serves to diminish both intraoperative and postoperative blood loss by preventing egress of blood. Although this theory has gained informal acceptance in orthopedic practice, scientific support remains sparse.21

This study was designed to assess comparative blood loss and transfusions in cemented/hybrid and noncemented THA by evaluating 25 matched pairs. Perioperative blood loss and transfusion requirements were compared between the two groups. The results have implications for transfusion requirement decisions in the future.

MATERIALS AND METHODS

Twenty-five noncemented THA patients were matched with 25 cemented/hybrid THA patients for comparison in a retrospective analysis. Patients were matched for age, gender, and weight, and carried the diagnosis of non-inflammatory arthritis. All surgery was performed at the same institution by two senior surgeons. Anesthesia (general, hypotensive) and operative approach (posterolateral) were standardized. The prostheses used were the Harris-Galante porous-coated hip for noncemented procedures (Zimmer, Inc., Warsaw, IN) and the HarrisGalante precoat hip for hybrid procedures. Deep vein thrombosis prophylaxis consisted of compression device only, with no pharmacologic interventions.

Intraoperative blood loss was measured by the volume of blood collected in the suction device and by the anesthesiologist's estimations. Postoperative blood loss was determined by output into a suction device left in for 24 hours. Comparisons were also made between hematocrit (HCT) and hemoglobin (Hb) preoperatively, immediately postoperatively, and on the day of hospital discharge. Operative time and length of stay were measured and compared. Statistical methods used to analyze data included the Student's / test, Mann-Whitney u test, KruskalWallis test, and the chi-square test.

RESULTS

The average age of patients in the cemented/hybrid group was 65.5 years compared to 64.0 years in the noncemented population. Operative times were equal; cemented/hybrid procedures lasted an average of 150 minutes compared to 148 minutes for noncemented. The relationship between operating room times and blood loss verifies that cemented/hybrid and noncemented procedures did not result in operating room time differences among this matched-pair population. Operating room blood loss averaged 1078 mL (cemented/hybrid) and 1207 mL (noncemented).

No significant differences were noted in autologous donation volumes (cemented/hybrid 1340 mL, noncemented 1371 mL) with an average autologous transfusion volume of 1 191 mL in the cemented/hybrid group and 1308 mL in the noncemented group (Table 1). Preoperative Hb levels averaged 12.5 g/dL in the cemented/hybrid group and 12.0 g/dL in the noncemented group (Table 2). Immediate postoperative Hb levels were 11.0 g/dL and 11.8 g/dL, respectively. Hemoglobin on the last day of hospitalization averaged 10.6 g/dL in the cemented/hybrid group and 11.1 g/dL in the noncemented group (Table 2).

Comparisons of groups did not demonstrate any significant differences in intraoperative and postoperative blood loss or in the volume of autologous or allogeneic transfusion (Table 1). However, when the total population was stratified by gender, the male population demonstrated a significantly greater blood loss (1848 mL) than the female population (1464 mL) (P < .004) (Table 3). There was no correlation between weight and blood loss among any group. Further stratification by gender demonstrated that a female subpopulation of 10 patients who were taking Premarin (conjugated estrogens; Wyeth-Ayerst, Philadelphia, PA) bled less in the operating room (810 mL versus 1209 mL) than the 22 females who were not (P < .0107). There were no differences in drain output (413 mL and 416 mL, respectively). This same population received 1 164.5 mL of transfused blood, which was less than the population of 22 females not on Premarin who were transfused (1376.7 mL) (Table 4).

Other significant findings include a shorter length of stay for those patients in the cemented/hybrid group (6.8 days versus 7.5 days) and a higher drain output for cemented/hybrid patients (545 mL versus 378 mL) (Table 1).

Table

TABLE 2Hemoglobin profiles

TABLE 2

Hemoglobin profiles

DISCUSSION

The threat of blood-borne infection has been one impetus for vigorous efforts to minimize allogeneic transfusions. u Total hip arthroplasty has been associated with blood loss often resulting in transfusion and the subsequent need to identify and implement blood conservation techniques.9·20 The purpose of this study was to directly compare blood loss and transfusion demands in noncemented and cemented/hybrid THA patients with the finding of no significant differences in perioperative blood loss or transfusion rates between the two groups.

A variety of measures have been used to minimize allogeneic transfusion for these surgical procedures. Perceived differences in relative risk of allogeneic transfusions have led some surgeons to apply different transfusion strategies in patients undergoing noncemented and cemented/hybrid THA. This study has placed these practices in question by not supporting the assumption that cemented/hybrid procedures result in a lesser surgical blood loss.

Predicting patients who are likely to require perioperative allogeneic transfusions has proven difficult.12 Part of the difficulty stems from the lack of accurate predictors regarding transfusion decisions. Historically, one common guideline was the "10/30" rule, suggesting that a Hb < 10 g/dL or an HCT < 30% was indicative of the need for transfused blood. These values were first challenged in the 1960s with evidence suggesting that Hb and HCT levels alone are insufficient measures for assessing the need for transfused blood.19 The selection of other reliable indicators of perioperative transfusion in arthroplasty patients has proven more difficult. Revision arthroplasty adds another factor, as revision procedures have been associated with increased blood loss and transfusion risk.13 The choice of procedure type was previously thought to be another predictor of perioperative blood loss and transfusion risks; however, results from the 25 matched-pair study do not support this contention.

Table

TABLE 3Gender blood loss

TABLE 3

Gender blood loss

One limitation of this study is the relatively small sample size (25 patients per group). However, patients were matched for age, gender, body weight, and diagnosis.

Prior studies have sought to demonstrate differences in blood loss and transfusion rates of cemented/hybrid and noncemented procedures. An et al21 studied 140 consecutive THA patients and demonstrated a significant difference in postoperative blood loss between patients who underwent cemented and noncemented procedures. Interestingly, while intraoperative blood loss was greater in the cemented group, postoperative blood loss was greater in the noncemented group. The An et al study also noted an increased rate of transfusion in the noncemented group, as well as an insignificant drop in Hb over the cemented group.

Table

TABLE 4Hormone replacement therapy population blood loss and transfusion rates

TABLE 4

Hormone replacement therapy population blood loss and transfusion rates

Similarly, in this 25 matched-pair study, intraoperative blood loss was greater in the cemented/hybrid cohort while postoperative drain output was greater in the noncemented group. Transfused blood (both autologous and allogeneic) was also greater in the noncemented group compared to the cemented/hybrid group. Although these differences were not statistically significant, they seem to parallel those found in the preceding study, strengthening the timing parameter of blood loss in THA patients.

CONCLUSION

No significant difference was found in perioperative blood loss and transfusion rates of noncemented and cemented/hybrid THA. The data suggest that there is no indication for different strategies in the perioperative blood management of these patients regarding relative transfusion risk. The results of this matched-pair study demonstrate a significant difference in blood loss between male and female patient populations, suggesting that autologous blood donation and other perioperative methods of decreasing transfusion risk should be evaluated and possibly modified on the basis of gender.

REFERENCES

1. Peterman TA, Jaffe HW, Feorino PM, et al. Transfusion-associated acquired immunodeficiency syndrome in the United States. JAMA. 1985; 254:2913-2917.

2. Royal College of Physicians of Edinburgh. Consensus Conference on Autologous Transfusion. Transfusion. 1996; 37:625-667.

3. Grzeskiewicz JL, Hall RA, Anderson G, Mintz PD. Preoperative crossmatch guidelines for total hip arthroplasty. Orthopedics. 1989; 12(5):549-553.

4. Desmond MJ, Thomas MJ, Gillon J. Fox MA. Consensus Conference on Autologous Transfusion. Perioperative red cell salvage. Transfusion. 1996; 36:644-651.

5. Toy PT, Beattie C. Gould SA, et al. Transfusion alert: Use of autologous blood. National Heart. Lung, and Blood Institute. 1994; 94:1-11.

6. Thomson JD, Callaghan JJ, Savory CG. Stanton RP, Pierce RN. Prior deposition of autologous blood in elective orthopaedic surgery. J Bone Joint Surg. 1987; 69:320-324.

7. Wasman J, Goodnough LT. Autologous blood donation for elective surgery. Effect on physician transfusion behavior. JAMA. 1987; 258:3135-3137.

8. Goodnough LT. Autologous blood donation [letter], JAMA. 1988; 259:2404-2405.

9. Goodnough LT, Wasman J, Corlucci K, Chernosky A. Limitations to donating adequate autologous blood prior to elective orthopedic surgery. Arch Surg. 1989; 124:494-496.

10. Ness PM, Bourke DL, Walsh PC. A randomized trial of perioperative hemodilution versus transfusion of preoperatively deposited autologous blood in elective surgery. Transfusion. 1992;31:226-230.

11. Oishi CS, D'Lüna DD, Morris BA. et al. Hemodilution with other blood reinfusion techniques in total hip arthroplasty. Clin Orthop. 1997;339:132-139.

12. Nelson CL, Fontenot HJ. Ten strategies to reduce blood loss in orthopedic surgery. Am J Surg. 1995; I70(suppl 6A):64S-68S.

13. Law JK, Wiedel JD. Autotransfusion in revision total hip arthroplasties using uncemented prostheses. Clin Orthop. 1989; 245:145-149.

14. Semkiw LB, Schurman DJ, Goodman SB, Woolson ST. Postoperative blood salvage using die Cell Saver after total joint arthroplasty. J Bone Joint Surg Am. 1989; 71 A:823-827.

15. Mallory TH. Hypotensive anesthesia in total hip replacement. JAMA. 1973; 224:248.

16. Davis NJ. Jennings JJ. Harris WH. Induced hypotensive anestesia for total hip replacement. Clin Orthop. 1974; 101:93-98

17. Reilly TJ. Gradisar LA Jr, Pakan W. Reilly M. The use of postoperative suction drainage in total knee arthroplasty. Clin Orthop. 1986; 208:238-242.

18. Amrein PC. Ellman L, Harris WH. Aspirin-induced prolongation of bleeding time and perioperative blood loss. JAMA. 1981; 245:1825-1828.

19. Nuttall GA, Santrach PJ, Oliver WC Jr. et al. The predictors of red cell transfusions in total hip arthroplasties. Transfusion. 1996; 36:144149.

20. Eckardt JJ, Gossett TC, Amstutz HC. Autologous transfusion and total hip arthroplasty. Clin Orthop. 1978; 132:39-45.

21. An HS. Mikhail WE, Jackson WT, Tolin B, Dodd GA. Effects of hypotensive anesthesia, nonsteroidal antiinflammatory drugs, and polymetfrylmethacrylate on bleeding in total hip arthroplasty patients. J Arthroplasty. 1991; 6:245-250.

TABLE 1

Postoperative outcomes

TABLE 2

Hemoglobin profiles

TABLE 3

Gender blood loss

TABLE 4

Hormone replacement therapy population blood loss and transfusion rates

10.3928/0147-7447-19990102-07

Sign up to receive

Journal E-contents