Blood Transfusion in Elective Abdominal Gynecologic Surgery

Abstract

Objective:

The aims of this research were to investigate the rate of blood use in elective gynecologic surgery, and to determine the appropriateness of blood ordering and utilization, and the factors associated with blood transfusion.

Materials and Methods:

This prospective study was conducted in women scheduled to undergo elective gynecologic surgery. Various surgery-related data were collected, including blood request data and actual blood use. Incidence of blood transfusion and other indices were estimated, including crossmatch-to-transfusion ratio (C/T ratio), transfusion index (TI), and transfusion probability (%T).

Results:

Two hundred and sixty women with a mean age of 48.5 ± 11.0 years were included. A majority of cases (42.7%) had benign uterine pathology, and 27.3% of cases had gynecologic malignancies. Blood requests were type and screen in 60.8% of cases. Almost half of all operations were performed by staff members (46.9%), and hysterectomy was the most common procedure (62.3%). The overall rate of blood use was 11.5%. The rate was 5.1% among type and screen cases, and 21.6% in crossmatching cases. Blood use was significantly more common among gynecologic malignancy operations than other procedures. Overall, the C/T ratio was 4.27, the TI was 0.23, and the %T was 23%, indicating over-ordering of blood. Inappropriate blood ordering was observed for patients with uterine pathology and gynecologic malignancies, for patients operated on by a fellow or staff member, and for patients undergoing hysterectomy with/without adnexal surgery or oncologic procedures.

Conclusions:

The rate of blood use was 11.5% in elective abdominal gynecologic surgery. Various indices showed that blood ordering and utilization is still inefficient and needs some improvements. (J GYNECOL SURG 33:231)

Introduction

Transfusion of blood is often a lifesaving measure in surgical patients; however, blood transfusion is uncommon in most operative procedures. Blood units are routinely ordered before elective surgery, and they are often not used during the procedure. Once ordered, this blood is held in reserve and is unavailable to other patients who might be in urgent need of transfusion. This blood management scheme is subject to blood storage problems in the blood bank, loss of shelf life, and waste of blood resources. The need for blood in hospitals continues to exceed the volume collected by blood collection services. Several studies have shown that blood ordered for elective surgical procedures often exceeds both actual and anticipated needs.^1–7

There are two standard methods used to type blood, namely type and screen and crossmatching. The type and screen method types blood using the ABO group system. In this system, the serum is saved so that further blood typing can be performed, if necessary. Type and screen is easier and faster to perform than crossmatching and does not require removal of blood from the common pool. The crossmatching method involves fully typing a sample and a unit of red cells to observe for crossreactivity. Crossmatched blood is ready to use, but it is removed from the common pool.

Transfusion guidelines and strategies designed to improve blood utilization have been shown to be cost-effective and safe.^8,9 Type and screen and crossmatching are both currently used at Siriraj Hospital, of Mahidol University, in Bangkoknoi, Bangkok, Thailand. There is, however, no established evidence-based guideline for various surgical procedures, including for gynecologic surgery. A previous study in Thailand reported overordering of blood for elective breast surgery, with a very high crossmatch-to-transfusion ratio (C/T) ratio of 29.1.¹⁰ However, information on blood use in elective gynecologic surgery and the efficacy of current blood-ordering practice in Thailand remains limited.

The aim of this study was to investigate the rate of blood use in elective gynecologic surgery, and to determine the appropriateness of blood ordering and utilization, and the factors associated with blood transfusion.

Materials and Methods

After receiving approval from the Siriraj institutional review board, a prospective study was conducted in the Department of Obstetrics and Gynecology, of the Faculty of Medicine at Siriraj Hospital. This hospital is Thailand's largest university-based national tertiary referral center. Two hundred and sixty women scheduled to undergo elective gynecologic surgery were included in the study. Patients with histories of anticoagulant drug use, coagulopathy, or repeated abdominal gynecologic surgery during the same admission were excluded.

After written informed consent was obtained, relevant information was recorded for each patient, including baseline characteristics, underlying disease, history of previous surgery, preoperative diagnosis, number of crossmatched blood orders requested, blood transfusions during the operation and until 24 hours post-operation, surgeon skill level, operative procedure, anesthetic method, operative time, estimated blood loss (EBL), and injury to other organs. Various transfusion indices were estimated using the formulas shown in Table 1.¹¹ This was an observational study in which cases were not randomly assigned to surgeons with different skill levels and the surgeons were not blinded to types of blood requests.

Table 1.

Calculation of Transfusion Indices ^a

Indices	Formula	Appropriate value
C/T ratio	# of units crossmatched # of units transfused	< 2.5
%T	# of patients transfused × 100 # of patients crossmatched	< 30%
TI	# of units transfused # of patients crossmatched	< 0.5

See Ref. 11.

C/T ratio, crossmatch-to-transfusion ratio; %T, transfusion probability; TI, transfusion index.

Data were analyzed using descriptive statistics. Data are presented as number and percentage, mean ± standard deviation, or median and interquartile range. Incidence of blood transfusion, C/T ratio, transfusion index (TI), and transfusion probability (%T) were estimated. Factors associated with blood transfusion were analyzed using a χ² test. A p-value of <0.05 was considered to be statistically significant.

Results

A total of 260 patients were included. Baseline clinical characteristics are shown in Table 2. Mean age was 48.5 years, mean body mass index was 24 kg/m², and mean hematocrit was 37.4%. Forty percent of patients had had previous abdominal surgery. A majority of cases were diagnosed with uterine pathology (42.7%), followed by gynecologic malignancy in 27.3%, adnexal pathology in 21.9%, and uterine with adnexal pathology in 8.1%.

Table 2.

Baseline Clinical Characteristics (N = 260)

Characteristics	Mean ± SD
Age (years)	48.5 ± 11.0
BMI (kg/m²)	24.6 ± 5.1
Hematocrit (%)	37.4 ± 4.1
	n (%)
Previous surgery	104 (40%)
Preoperative diagnosis
Uterine pathology	111 (42.7%)
Adnexal pathology	57 (21.9%)
Uterine with adnexal pathology	21 (8.1%)
Gynecologic malignancy	71 (27.3%)

SD, standard deviation; BMI, body mass index.

Blood requests and blood use details are given in Table 3. Blood orders were type and screen in 60.8% of cases, and crossmatching in 39.2%. The number of units requested was 2 units in 74.5% of cases, and 1 unit in 25.5%. Blood transfusion occurred in 30 cases (11.5%), and a majority of cases used 1 or 2 units (86.6%).

Table 3.

Blood Requests & Blood Use Among Study Population (N = 260)

Blood request & blood use	N (%)
Type of blood request
Type & screen	158 (60.8%)
Crossmatching	102 (39.2%)
Amount of blood request
1 unit	26 (25.5%)
≥ 2 units	76 (74.5%)
Actual blood use (cases)	30 (11.5%)
Amount of blood use (units)
1 unit	13 (43.3%)
2 units	13 (43.3%)
≥ 3 units	4 (3.4%)

Most elective abdominal gynecologic surgery was performed by staff members (46.9%). The most common operation was hysterectomy with or without adnexal surgery (62.3%). Median operative time was 130 minutes, and median EBL was 250 mL (Table 4).

Table 4.

Operative Characteristics of Enrolled Patients (N = 260)

Operative characteristics	n (%)
Surgeon skill level
Resident	72 (27.7%)
Fellow	66 (25.4%)
Staff	122 (46.9%)
Procedure
Hysterectomy with/without adnexal surgery	162 (62.3%)
Adnexal surgery	16 (6.2%)
Oncologic procedure	82 (31.5%)
	Median (IQR)
Operative time	130 (95, 176.75)
EBL	250 (100, 500)

IQR, interquartile range; EBL, estimated blood loss.

A comparison between characteristics and type of blood requests is presented in Table 5. Crossmatching was significantly more common in gynecologic malignancy cases than in other diagnoses (p < 0.001). Crossmatching was also significantly more common among cases in which the surgeon was a surgical fellow than among cases operated by the other two types of surgeons. There was no difference between the two blood typing methods relative to history of previous surgery (Table 5).

Table 5.

Comparison Between Characteristics & Types of Blood Requests

Characteristics	Type and screen (n = 158)	Crossmatching (n = 102)	p-Value
Preoperative diagnosis			< 0.001
Uterine pathology	96 (86.5%)	15 (13.5%)
Adnexal pathology	33 (57.9%)	24 (42.1%)
Uterine & adnexal pathology	15 (71.4%)	6 (28.6%)
Gynecologic malignancy	14 (19.7%)	57 (80.3%)
Surgeon skill level			< 0.001
Resident	60 (83.3%)	12 (16.7%)
Fellow	8 (12.1%)	58 (87.9%)
Staff	90 (73.8%)	32 (26.2%)
Previous surgery	68 (65.4%)	36 (34.6%)	0.213

Table 6 shows a comparison between characteristics and blood use. Blood use was significantly more common for patients with crossmatching blood orders than type and screen blood orders (p < 0.001). Although the difference did not achieve statistical significance, blood use was greater for cases in which the surgeon was a fellow (p = 0.053). Patients undergoing oncologic procedures were significantly more likely to receive blood transfusion than patients undergoing other procedures (p = 0.014). No patients who underwent adnexal surgery in this study required blood transfusions.

Table 6.

Comparison Between Characteristics & Blood Use

Characteristics	Blood use n (%)	p-Value
Blood request type		<0.001
Type & screen (n = 158)	9 (5.7%)
Crossmatching (n = 102)	21 (21.6%)
Surgeon skill level		0.053
Resident (n = 72)	7 (9.7%)
Fellow (n = 66)	13 (19.7%)
Staff (n = 122)	10 (8.2%)
Procedure		0.014
Hysterectomy with/without adnexal surgery (n = 162)	14 (8.6%)
Adnexal surgery (n = 16)	0 (0%)
Oncologic procedure (n = 82)	16 (19.5%)

With respect to transfusion indices, the overall C/T ratio was 4.27, %T was 23%, and TI was 0.23. While the C/T ratio was above the recommended value, %T and TI were both at acceptable levels.

A comparison between characteristics and blood transfusion indices is given in Table 7. CT ratio was >2.5 for patients diagnosed with uterine pathology, diagnosed with gynecologic malignancy, operated on by a surgical fellow, undergoing hysterectomy with/without adnexal surgery, and undergoing oncologic procedure. Transfusion probability was >30% in patients diagnosed with uterine with adnexal pathology, and in patients operated on by residents. TI was >0.5 only among patients diagnosed with uterine with adnexal pathology.

Table 7.

Comparison Between Characteristics & Blood Transfusion Indices

Characteristics	C/T ratio	%T	TI
Preoperative diagnosis
Uterine pathology	5.5^a	20.0%	0.18
Adnexal pathology	N/A	0.0%	0.0
Uterine with adnexal pathology	1.13	66.7%^a	0.89^a
Gynecologic malignancy	4.1^a	24.2%	0.24
Surgeon skill level
Resident	2.0	33.3%^a	0.5
Fellow	4.3^a	23.3%	0.23
Staff	5.7^a	20.0%	0.18
Procedure
Hysterectomy with/without adnexal surgery	4.1^a	24.1%	0.24
Adnexal surgery	N/A	0.0%	0.0
Oncologic procedure	4.1^a	24.2%	0.24

Values above the appropriate level.

C/T ratio, crossmatch-to-transfusion ratio; %T, transfusion probability; TI, transfusion index; N/A, not available.

Discussion

Blood transfusions play a major role in the management of surgical patients, and blood ordering is a common practice in elective and emergency surgical procedures. However, the ordering and use of blood must be optimized in order to maximize the availability of blood resources and to minimize costs. Variations in transfusion rates are due to many factors, including differences in case mix, surgical and anesthetic techniques, preoperative patient conditions, and lack of transfusion protocols.

This study revealed that 11.5% of patients who underwent elective gynecologic surgery at the current authors' center received blood transfusions. Significant procedures associated with blood transfusions were oncologic procedures in gynecologic malignancy cases (19.5%). This rate was higher than the 13.8% rate reported in a 2015 study.¹² In the present study, patients operated on by surgical fellows received blood transfusions more commonly. This might, in part, be explained by the fact that patients operated on by fellows were more often oncologic cases. Without an established guideline, blood ordering is usually based on subjective anticipation of blood loss for a surgical procedure. As such, surgeons order blood prepared by the crossmatching technique more commonly. This was demonstrated in the present study by blood transfusion being more common among crossmatched cases.

Regarding the appropriateness of blood ordering and its use, over-ordering of blood was demonstrated in the present study by a high C/T ratio and low %T and TI. The results showed an overall C/T ratio of 4.27, which is greater than the recommended value of 2.5. This means that only 23.4% of crossmatched blood units were transfused. The %T of 23% and TI of 0.23 suggest that blood use among this patient population is not significant. It has been suggested that procedures with a TI value of <0.5 may not require preoperative crossmatching.^11,13 A previous study conducted in Thailand reported overordering of blood for elective breast surgery based on a very high C/T ratio of 29.1.¹⁰

The C/T ratio of 4.27 in the present study was relatively higher than those previously reported from other studies with similar settings. A recent study in a tertiary-care center in India reported an overall C/T ratio of 1.6 and this ratio was highest in an obstetrics and gynecology unit, followed by surgery units (C/T ratios of 2.7 and 2.1, respectively).¹⁴ Another study from a tertiary-care teaching hospital in India reported an even higher CT ratio of 4.82 in an obstetrics and gynecology unit with %T of 41.26% and TI of 4.81. Detailed analysis showed C/T ratios of 4.34 and 4.81 in patients diagnosed with uterine myomas and ovarian cystadenomas, respectively.¹⁵ Earlier reports from Africa reported overall C/T ratios of 2.2 and 1.64 in an obstetrics and gynecology unit in a teaching university hospital in Nigeria.¹⁶ In another study from Ethiopia, the overall C/T ratio, %T, and TI were 2.3, 47%, and 0.77, respectively, and in an obstetrics and gynecology unit, a higher C/T ratio was observed among elective, compared to emergency cases (2.9 and 1.8, respectively).⁷ However, these previous studies included both obstetric and gynecologic cases together in their analyses; thus, comparisons with the results of this study should be made with caution.

Over-ordering of blood was observed for patients with either a preoperative diagnosis of uterine pathology or a preoperative diagnosis of gynecologic malignancy, patients operated on by a fellow or a staff surgeon, and patients undergoing either hysterectomy with or without adnexal surgery or oncologic procedures. There might be excessive anticipation of blood loss for these patients. In contrast, significant blood use was found for patients with preoperative diagnoses of uterine with adnexal pathology and in patients operated on by surgical residents. High blood use in this setting could be explained by cases that are more complicated than expected involving patients who are operated on by residents with less surgical experience and lower operative skills. Inappropriate utilization of blood could be decreased by simply changing the pattern of blood typing and ordering in relation to the type of surgery being performed.^3,17 Other strategies include timely and adequate preoperative assessment of risk, anticipation of potential transfusion problems, and intraoperative techniques to minimize blood loss.¹⁸

The use of a Maximal Surgical Blood Order Schedule (MSBOS) has been proposed and has succeeded in limiting unnecessary transfusion practices.⁸ The MSBOS is used to estimate the amount of blood that will be needed for a specific procedure. The estimated number of crossmatched units is based on institutional usage statistics for a given surgical procedure.¹⁹ The MSBOS has been shown to save both resources and costs in blood ordering and transfusion practice.⁹ Several studies have confirmed that over-ordering of blood exists in various fields of surgical practice, and that each institution should create MSBOS protocols based on blood usage rates at their centers for specific surgical procedures.^3–7,17 In addition, a Patient-Specific Blood Ordering System has been suggested for use as an additional tool for predicting patients who are at risk for receiving blood transfusions during surgery.^20,21

A previous study in Thailand reported that the C/T ratio was reduced from 3.6 to 2.6 after implementation of a blood ordering guideline for elective surgery.²² An earlier study in India revealed that unutilized blood was significantly reduced from 76.86% to 25.26% after applying a new blood ordering schedule based on various transfusion indices.³ A study in a university hospital in Pakistan reported a very high C/T ratio of 20.27, probably due to the routine use of a crossmatching blood ordering system for elective surgery.¹¹ Another study in Malaysia revealed a significant reduction in unnecessary or inappropriate crossmatching of blood for routine elective surgical cases after introduction of an MBOS and type and screen procedure.⁵ A study in an academic medical center in the United States reported a significant decrease in percentage of procedures with preoperative blood orders from 40.4% to 25.0% and also a significant decrease in C/T ratio from 2.11 to 1.54, after an MBOS was introduced.²³

In the absence of an explicit MSBOS, blood ordering for transfusion is frequently based on subjective anticipation of blood loss instead of audit-based requirement estimates for a particular procedure. As a consequence, suboptimal blood ordering and utilization are likely to continue. Once an MSBOS is established, regular audits are necessary to ensure that changes in blood ordering and use are accounted for and that the protocol for that procedure is current. In addition to the use of procedure-specific blood use statistics, input from and agreement among affected surgeons should be factored into the development of an MSBOS. Provisions that allow for flexibility in blood ordering protocols are also needed to accommodate cases in which anticipated blood losses are greater due to patient and operative factors so that patient safety is never compromised.

This study had some limitations. First, the case mix, relative to preoperative status and operative difficulties, might have partially affected the results. Second, although the results were stratified by surgeon skill level, some variations in experience, skill level, and specific expertise still exist within each category. Third, given that this study was conducted in a large national tertiary referral hospital, enrolled cases may have been more complicated than those that might be found in other settings. As such, the results of this study may not be generalizable to patient populations in other settings within Thailand.

Conclusions

The rate of blood use was 11.5% in elective abdominal gynecologic surgery. Significant blood use was found for patients with preoperative diagnoses of uterine with adnexal pathology, and for patients operated on by surgical residents. Given the inefficiencies found for blood ordering and utilization, the findings from this study can be used to develop a blood ordering guideline or MSBOS for elective gynecologic surgery at the current authors' center.

Footnotes

Author Disclosure Statement

The authors declare that no conflicts of interest exist.

References

Friedman

, Oberman

, Chadwick

, Kingdon

. The maximum surgical blood order schedule and surgical blood use in the United States. Transfusion, 1976; 16:380.

Hall

, Pattenden

, Hollobone

, Pollard

, Dennison

. Blood transfusion policies in elective general surgery: How to optimise cross-match-to-transfusion ratios. Transfus Med Hemother, 2013; 40:27.

Vibhute

, Kamath

, Shetty

. Blood utilisation in elective general surgery cases: Requirements, ordering and transfusion practices. J Postgrad Med, 2000; 46:13.

Chawla

, Kakepoto

, Khan

. An audit of blood cross-match ordering practices at the Aga Khan University Hospital: First step towards a maximum surgical blood ordering schedule. JPMA J Pak Med Assoc, 2001; 51:251.

Jayaranee

, Prathiba

, Vasanthi

, Lopez

. An analysis of blood utilization for elective surgery in a tertiary medical centre in Malaysia. Malays J Pathol, 2002; 24:59.

Singh

, Singh

. Routine pre-operative cross-match for elective colorectal resections: An appropriate use of resources?. Surgeon, 2011; 9:8.

Belayneh

, Messele

, Abdissa

, Tegene

. Blood requisition and utilization practice in surgical patients at University of Gondar Hospital, Northwest Ethiopia. J Blood Transfus, 2013; 2013:758910.

Murphy

, Phillips

, Gray

, et al. Blood use for surgical patients: A study of Scottish hospital transfusion practices. J R Coll Surg Edinb, 1995; 40:10.

Richardson

, Bradley

, Donaldson

, O'Shaughnessy

. Maximum surgical blood ordering schedule in a district general hospital saves money and resources. Ann R Coll of Surg Eng, 1998; 80:262.

10.

Litu

, Simajareuk

, Chau-in

, Boonsangcharoen

, Vachirodom

. Pre-operative routine cross-match for elective breast surgery: An appropriate use of resources. Srinagarind Med J, 2012; 27:401.

11.

Soomro

, Javed

, Ali

. Arrangements and use of blood in elective surgical procedures. Professional Med J, 2011; 18:212.

12.

Prescott

, Aloia

, Brown

, et al. Perioperative blood transfusion in gynecologic oncology surgery: Analysis of the National Surgical Quality Improvement Program Database. Gynecol Oncol, 2015; 136:65.

13.

Mead

, Anthony

, Sattler

. Hemotherapy in elective surgery: An incidence report, review of the literature, and alternatives for guideline appraisal. Am J Clin Pathol, 1980; 74:223.

14.

Kumari

. Blood transfusion practices in a tertiary care center in Northern India. J Lab Physicians, 2017; 9:71.

15.

Raghuwanshi

, Pehlajani

, Sinha

, Tripathy

. A retrospective study of transfusion practices in a tertiary care institute. Indian J Anaesth, 2017; 61:24.

16.

Ebose

, Osalumese

. Blood shortage situation: An audit of red blood cells order and pattern of utilization Afr. J Biotechnol, 2009; 8:5922.

17.

de Jongh

, Feng

, Frank

, Wallace

. Improved utilization of blood for elective surgery. Surg Gynecol Obstet, 1983; 156:326.

18.

Sherman

, Macivor

. Blood utilization: Fostering an effective hospital transfusion culture. J Clin Anesth, 2012; 24:155.

19.

Dodsworth

, Dudley

. Increased efficiency of transfusion practice in routine surgery using pre-operative antibody screening and selective ordering with an abbreviated crossmatch. Br J Surg, 1985; 72:102.

20.

Palmer

, Wahr

, O'Reilly

, Greenfield

. Reducing unnecessary cross-matching: A patient-specific blood ordering system is more accurate in predicting who will receive a blood transfusion than the maximum blood ordering system. Anesth Analg, 2003; 96:369.

21.

Karger

, Kretschmer

. Evaluation of blood ordering algorithms. Anesth Analg, 2003; 97:927.

22.

Mahattanobon

, Sunpaweravong

. Blood order guideline for elective surgery: Impact of a guideline. Songkla Med J, 2008; 26:491.

23.

Frank

, Oleyar

, Ness

, Tobian

. Reducing unnecessary preoperative blood orders and costs by implementing an updated institution-specific maximum surgical blood order schedule and a remote electronic blood release system. Anesthesiology, 2014; 121:501.