Approximately 31 million women in the United States suffer from pelvic floor disorders (PFDs), resulting in approximately 376,700 surgeries for pelvic organ prolapse (POP) and urinary incontinence annually.1,2 Red blood cell (RBC) transfusion rates with PFD surgery have been reported between 0.2% and 16%,3,4 leading many surgeons to order a preoperative blood type and antibody screen (preoperative type and screen [T&S]).
To reduce the risk of transfusion error, the College of American Pathologists in May 2015 issued revised guidelines for T&S that now require verification of a patient's blood type from a separate blood draw for a T&S to be valid.5 This additional testing may increase patients' discomfort, as well as cost to the health care system, and it raises an important clinical question: “Is it worthwhile to obtain a preoperative T&S prior to PFD surgery?” Answering this question requires information for this surgical population regarding (1) the current rate of transfusion with PFD surgery and (2) the risk of a hemolytic transfusion reaction due to unidentified antibodies if transfusion is needed and cross-matched blood is not available.
The goals of this study are to provide baseline information about (1) the likelihood of blood transfusion via various routes of POP surgery, (2) the likelihood of identifying a positive antibody with preoperative T&S in a urogynecologic surgical population, and (3) the expected frequency of antibody-mediated transfusion reaction if preoperative T&S is not ordered. We focused on reporting this information for apical POP procedures, as we felt these procedures represent complex POP surgery, which may have a high risk of blood transfusion. Thus, our primary objective was to compare perioperative blood transfusion rates between 3 routes of apical prolapse surgery (abdominal, robotic, vaginal). Our secondary objective was to report the prevalence of a positive antibody screen in women who underwent apical POP surgery.
MATERIALS AND METHODS
After institutional review board approval at the University of North Carolina at Chapel Hill (IRB 16-1626) and in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines,6 we conducted a retrospective cohort study of women 18 years or older who underwent apical POP surgery by any surgeon in the Division of Female Pelvic Medicine and Reconstructive Surgery between May 2005 and May 2016. Operative records were reviewed to identify the most recent, consecutive apical POP surgeries until we reached our estimated sample size. We divided our study population into 3 groups based on route of apical POP repair. The abdominal group was composed of women who underwent abdominal sacrocolpopexy (CPT 57280), the robotic group was composed of women who had a robotic sacrocolpopexy (CPT 57425), and the vaginal group was composed of women who underwent vaginal uterosacral (CPT 57282) or sacrospinous ligament suspension (CPT 57283). Surgeries undergoing conversion from one route to another (eg, robotic to abdominal) were analyzed according to the initially attempted route. Women with prior POP surgery were included. Women were excluded if they had a known bleeding disorder, were taking anticoagulant medication, or underwent combined surgery with another surgical service at the time of POP repair. Patients undergoing colpocleisis were not included because we felt that the obliterative nature of this procedure presented a fundamentally different surgical approach.
Our primary outcome was the rate of RBC transfusion during the primary hospital admission associated with the apical POP surgery compared by route of surgery. We determined the rate of transfusion by reviewing the electronic medical record to identify if RBC units were transfused intraoperatively and/or postoperatively during admission. Our secondary outcome was the prevalence of a positive antibody screen among the women who underwent a preoperative T&S.
We also abstracted data from the electronic medical record regarding age, race, parity, Charlson Comorbidity Index,7,8 surgical history, smoking status, body mass index (BMI), and Pelvic Organ Prolapse Quantification9 stage. We reviewed operative reports to determine estimated blood loss, concomitant procedures, and intraoperative complications including, bladder, bowel, or ureter injury. We also reviewed pathology reports to collect uterine weights for patients who underwent hysterectomy.
For our sample size estimate, we used published transfusion rates for PFD surgery, which ranged from 0.2% to 16%.3,4 We felt that a difference in transfusion rates of 10% for abdominal versus 3% for robotic and 3% for vaginal surgeries was clinically relevant and supported by the literature. Using a 1:1 allocation, with 80% power and an α of 0.05, we determined that 195 women were needed in each group to detect a transfusion rate difference of 10% in the abdominal group versus 3% in the robot, as well as 10% in the abdominal group versus 3% in the vaginal group.
Data were analyzed using SPSS 24 (IBM, Armonk, NY). To determine if there were demographic or clinical differences between women within the 3 groups (abdominal, robotic, vaginal), we used analysis of variance for continuous data, Kruskal-Wallis for nonparametric data, and χ2 or Fisher exact test for categorical data; furthermore, when making pairwise comparisons across 3 groups, we used a Bonferroni correction and set a significant P value as 0.0167 to correct for the increased chance of a type I error. To assess which variables were associated with our primary outcome (RBC transfusion), we performed a bivariate analysis using Student t, Mann-Whitney U, χ2, and Fisher exact tests as appropriate. We used multivariable logistic regression analysis to assess if transfusion was associated with route of surgery while adjusting for potential confounders.
Among the 610 women who met inclusion criteria, 199 (32.6%) underwent abdominal sacrocolpopexy, 207 (33.9%) underwent a robotic sacrocolpopexy, and 204 (33.4%) underwent a vaginal uterosacral or sacrospinous ligament suspension (Table 1). Of the 204 vaginal cases, 146 (71.5%) were uterosacral, and 58 (28.4%) were sacrospinous ligament suspensions. There were 4 procedures that were planned to be robotic that were converted to abdominal intraoperatively. None of these received transfusion. No vaginal cases were converted.
In comparing the 3 groups (abdominal, robotic, vaginal), differences were seen in age (P < 0.001), likelihood of white race (P = 0.007), and frequency of stages 3 to 4 POP (P = 0.001). Pairwise comparison of these outcomes showed women in the abdominal group were older compared with the robotic (P = 0.008) and vaginal (P < 0.001) groups, with no significant difference in age when comparing the robotic to the vaginal group (P = 0.09). White race was more prevalent in the abdominal group compared with vaginal (P = 0.003), whereas there was no significant difference in white race between abdominal and robot (P = 0.33) or robotic and vaginal (P = 0.04). Stages 3 to 4 prolapse was more common in the abdominal group compared with the robotic (P < 0.001) and vaginal (P < 0.001) groups, with no significant difference between the robotic and vaginal groups (P = 0.12). There were no differences between groups when comparing parity, Charlson Comorbidity Index, smoking status, BMI, intraoperative complications, or uterine weight.
Concomitant procedures varied by route of apical POP repair (Table 1). Concurrent hysterectomy was more common in the vaginal group compared with the abdominal (P < 0.001) and robotic (P < 0.001) groups. The likelihood of performing a concomitant incontinence procedure did not differ by route of surgery (P = 0.22); however, Burch colposuspension or fascial sling was more likely in the abdominal group compared with the robotic (P < 0.001) and vaginal (P < 0.001) groups, whereas midurethral sling was more common in the vaginal (P < 0.001) and robotic (P < 0.001) groups compared with the abdominal group. Anterior and posterior colporrhaphies were also more common in the vaginal group.
For our primary outcome, 24 women (3.9%) received an RBC transfusion. Transfusion was more common in women who underwent abdominal surgery 22 (11.1%) compared with robotic 1 (0.5%, P < 0.001) and vaginal 1 (0.5%, P < 0.001) surgery. Women who received a transfusion were older (P = 0.002), more likely to have stages 3 to 4 prolapse (P = 0.01), and less likely to undergo concurrent hysterectomy (P = 0.009) (Table 2). In a multivariable logistic regression model including age, stages 3 to 4 POP, concurrent hysterectomy, and route of surgery, transfusion remained associated with the abdominal route of surgery. This finding was consistent between 2 models that used either robotic (odds ratio [OR], 20.7; 95% confidence interval [CI], 2.7–156.6) or vaginal (OR, 18.7; 95% CI, 2.5–142.2) as the reference group for route of surgery. In addition to route of surgery, age in decades was significantly associated with transfusion (OR, 1.1; 95% CI, 1.0–1.1).
Of the 24 women (3.9%) who received a transfusion, 4 were transfused intraoperatively—all of whom were in the abdominal group. Two of these received 1 unit of RBC, 1 received 2 units, and 1 received 6 units. Twenty-two women were transfused postoperatively, of whom 20 were in the abdominal group, 1 in the robotic group, and 1 in the vaginal group (sacrospinous ligament fixation). Four of the women transfused received 1 unit of RBC, 15 received 2 units, and 3 received 3 units. There were no transfusion reactions in any of the recipients. To evaluate the impact of surgeon on transfusion rate, we evaluated the time period prior to the introduction of robotic surgery (as all surgeons performed abdominal and vaginal, but not all performed robotic surgery) in our study (2005–2010), and there were no differences in transfusion rates among surgeons, P = 0.25. When comparing the year that transfusion occurred, 17 transfusions occurred between 2005 and 2010, and 7 occurred from 2011 to 2016.
Among the 572 women (94.1%) who had a preoperative T&S ordered and performed preoperatively, 9 (1.5%) had a positive antibody screen. Only 7 of these (1.2%) were clinically significant, meaning that the antibodies detected were capable of causing hemolytic transfusion reaction or decreased RBC survival.10 The 7 clinically significant antibodies included 3 anti-E, 2 anti-K, 1 anti-C, and 1 anti-D. The 2 antibody screens that were considered nonsignificant were “warm” and “nonspecific” antibodies.
We observed significantly higher rates of transfusion for abdominal apical POP surgery compared with vaginal and robotic surgery in our study population. There are several possible explanations for why transfusion was more common in the abdominal route of apical POP repair. First, there were 4 intraoperative transfusions in the abdominal group, suggesting that intraoperative hemorrhage is more likely in this route. Second, concomitant incontinence procedures with abdominal cases were primarily open retropubic procedures, which may have an increased bleeding risk compared with the minimally invasive slings performed in the majority of robotic and vaginal surgeries in this study. Third, the magnification of blood vessels with robotic surgery may mitigate the risk of bleeding with robotic sacrocolpopexy, while not operating in the presacral space may explain why transfusion is less common with the vaginal approach. Importantly, there was no uniform transfusion protocol at our institution during the years studied, and the majority of the abdominal procedures occurred earlier in the study period and thus occurred when the transfusion thresholds may have been more liberal.
The findings of this study provide important baseline data to surgeons considering whether to order a preoperative T&S for women preparing for PFD surgery. Previously reported transfusion rates for abdominal PFD surgeries range from 6% to 16%,3,11,12 whereas transfusion rates for vaginal and robotic PFD surgeries range from 0.2% to 1.6%4,13 and 0.3% to 1.4%,12–14 respectively. Our findings of an 11.1% transfusion rate for abdominal and 0.5% for both robotic and vaginal surgeries fall within these reported ranges and confirm that the abdominal route is associated with higher rates of transfusion. Given the low transfusion rates for vaginal and robotic surgery, it may not be worthwhile to obtain a preoperative T&S for these routes of surgery.
To further evaluate the safety of not ordering a preoperative T&S, we need to consider what would occur if a patient does not have a preoperative T&S when an emergent transfusion is required. In this case, un–cross-matched type O blood can be used; however, there is a risk of an acute or delayed hemolytic transfusion reaction with un–cross-matched blood because of the possibility of undiagnosed alloantibodies. The frequency of a positive antibody screen in our study population was 1.5%; this is similar to that of the general female population, which is ~2%.15,16 Just as the rate of a positive antibody screen is low, the risk of a hemolytic transfusion reaction with type O blood is only 0.02% to 0.4% of emergency transfusions10,17 because an antibody-mediated transfusion reaction will occur only if a patient has an antibody to the same antigen that is present in the transfused blood. Therefore, un–cross-matched type O blood is considered very safe for use in emergency transfusions.
With the above information and our study results, we can estimate the likelihood of a hemolytic transfusion reaction for different routes of apical POP surgery. For example, if a preoperative T&S is not performed prior to an abdominal apical POP repair, the risk of hemolytic transfusion reaction is the risk of a transfusion (11.1%) multiplied by the rate of a transfusion reaction with type O blood (0.4%), which is 0.044% or ~1 in 2300 abdominal procedures. For robotic and vaginal apical POP repair, the risk of a hemolytic transfusion reaction is 0.002% (0.5% rate of transfusion × 0.4% risk of transfusion reaction) or ~1 in 50,000 robotic or vaginal procedures. Given this very low risk of hemolytic transfusion reaction with robotic and vaginal apical POP procedures, these findings support not ordering a preoperative T&S prior to uncomplicated robotic or vaginal apical POP surgery. The risks of non–antibody-mediated transfusion reactions such as transfusion-associated circulatory overload, transfusion-related acute lung injury, and anaphylactic transfusion reaction are independent of antibody status and thus not reduced by performing a preoperative T&S.
An additional consideration is the cost of a preoperative T&S to the health care system. At our institution, the change in the College of American Pathologist guidelines now requires patients who do not have a prior blood type on record to undergo a second venipuncture to confirm blood type before nonemergency release blood can be used. This can result in increased patient burden and cost. The cost of a blood type at our institution is $103, and antibody screen is $92. So, for a patient to undergo a preoperative T&S, including an additional venipuncture for a second blood type, the laboratory cost is $298. Thus, for every 1000 patients who undergo a preoperative T&S, the cost to the health care system is approximately $298,000. If these figures are extrapolated to other major gynecologic surgeries, the costs to the health care system are substantially higher. While a cost-benefit analysis is needed to explore the complications and other costs of hemolytic transfusion reaction as well as assess the implications of other possible outcomes such as an increased demand for O-negative blood that may occur if preoperative T&S is routinely deferred, the principle remains that in a time of limited health care resources these laboratory tests may represent a significant avoidable expense.
There are no widely adhered to US guidelines to suggest which gynecological surgical patients should undergo a preoperative T&S.18 However, literature exists in other surgical fields advocating for a selective approach to preoperative testing including a preoperative T&S.19,20 A review in the pediatric surgery literature concluded that a preoperative T&S was unnecessary in cases with a less than 5% transfusion probability.21 Many labor and delivery units in the United States have implemented a risk-based strategy to determine if T&S should be done upon admission. For example, in 2009, Stanford University's Labor and Delivery stratified patients by medical history as low risk of hemorrhage (managed with blood type only), moderate risk (T&S), and high risk (T&S plus cross-match).22 Implementation of these strategies resulted in a 27% reduction in diagnostic testing and 24% cost reduction23; however, the authors did not report patient outcomes in this retrospective study because of concern for bias.24 In this era of minimally invasive surgery where transfusion rates are low, risk-stratified preoperative testing may provide a safe and cost-saving alternative to routine ordering of a preoperative T&S. Some conditions that increase the risk of a positive antibody screen include a history of prior blood product transfusion or transfusion reaction, as well as medical comorbidities such as sickle cell disease, severe anemia, and solid organ malignancy.15,25,26 Thus, this history may warrant consideration of a preoperative T&S if risk of perioperative transfusion is not low.
The strengths of this study include adequate power to compare transfusion rates by route of apical POP surgery. Also, because we evaluated complex surgical procedures, the above worst-case scenario calculations likely exaggerate the calculated risks of hemolytic transfusion reaction when compared with surgeries with less potential for hemorrhage. In addition, the recent change in College of American Pathologist requirements for T&S prompts reconsideration of the utility of this preoperative laboratory test, making these findings timely.
This study is limited by its retrospective design. We minimized selection bias by examining consecutive surgeries among multiple surgeons in a large number of women. However, the generalizability of this study's findings may be limited given that it was conducted at a single academic center. Although multiple surgeons were included, not all surgeons utilized all 3 routes of apical surgery, and therefore an individual surgeon's surgical technique or transfusion threshold may have a greater impact on the transfusion rate in that surgeon's preferred route, and there was no standard policy during the study period to inform decision making for transfusion. In addition, although trainee involvement in cases was universal in this study, training level of the trainee was not recorded, and it is possible cases with lower training level assistants may have higher rates of transfusion. Another possibility is that transfusion was more common in the abdominal group because of the indications for abdominal surgery such as a history of prior abdominal surgery. Further, transfusion was more common in the early years of the study when robotic surgery was less common; thus, higher transfusion rates in the abdominal group may be overrepresented because of a difference in transfusion threshold over time. Finally, because 5.9% of women in this study did not have a preoperative T&S, it is possible that the positive antibody screen rate could have been higher; however, there were no significant demographic differences between those who did and did not undergo a preoperative T&S in our study. Furthermore, even if all of these patients had a positive antibody screen, the risk of hemolytic transfusion reaction with un–cross-matched blood would still be low.
Familiarity with the surgical likelihood of transfusion for PFD surgeries and the risk associated with the use of type O blood informs the need for a preoperative T&S in this population. Given the low risk of transfusion and even lower risk of antibody-mediated transfusion reaction in robotic and vaginal apical POP surgeries, not performing a preoperative T&S may be reasonable, cost saving, and associated with less patient burden. Future studies should include formal cost-benefit analysis to validate these findings.
1. Brueseke T, Muffly T, Rayburn W, et al. Workforce analysis of female pelvic medicine and reconstructive surgery, 2015 to 2045. Female Pelvic Med Reconstr Surg
2. Wu JM, Kawasaki A, Hundley AF, et al. Predicting the number of women who will undergo incontinence and prolapse surgery, 2010 to 2050. Am J Obstet Gynecol
3. Lambrou NC, Buller JL, Thompson JR, et al. Prevalence of perioperative complications among women undergoing reconstructive pelvic surgery. Am J Obstet Gynecol
4. Erekson EA, Yip SO, Ciarleglio MM, et al. Postoperative complications after gynecologic surgery. Obstet Gynecol
5. College of American Pathologists Commission on Laboratory Accreditation. Transfusion Medicine Checklist
. Northfield, IL; 2015. Available at: http://www.cap.org/ShowProperty?nodePath=/UCMCon/Contribution%20Folders/DctmContent/education/OnlineCourseContent/2016/LAP-TLTMv2/checklists/cl-trm.pdf
. Accessed April 16, 2018.
6. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol
7. Quan H, Li B, Couris CM, et al. Updating and validating the Charlson Comorbidity Index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol
8. Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis
9. Bump RC, Mattiasson A, Bø K, et al. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol
10. Mulay SB, Jaben EA, Johnson P, et al. Risks and adverse outcomes associated with emergency-release red blood cell transfusion. Transfusion
11. Stepp KJ, Barber MD, Yoo EH, et al. Incidence of perioperative complications of urogynecologic surgery in elderly women. Am J Obstet Gynecol
2005;192(5 spec. issue):1630–1636.
12. Geller EJ, Siddiqui NY, Wu JM, et al. Short-term outcomes of robotic sacrocolpopexy compared with abdominal sacrocolpopexy. Obstet Gynecol
13. Swenson CW, Kamdar NS, Harris JA, et al. Comparison of robotic and other minimally invasive routes of hysterectomy for benign indications. Am J Obstet Gynecol
14. Unger CA, Paraiso MF, Jelovsek JE, et al. Perioperative adverse events after minimally invasive abdominal sacrocolpopexy. Am J Obstet Gynecol
15. Saverimuttu J, Greenfield T, Rotenko I, et al. Implications for urgent transfusion of uncrossmatched blood in the emergency department: the prevalence of clinically significant red cell antibodies within different patient groups. Emerg Med
16. Geifman-Holtzman O, Wojtowycz M, Kosmas E, et al. Female alloimmunization with antibodies known to cause hemolytic disease. Obstet Gynecol
17. Goodell PP, Mohammed M, Powers AA. Risk of hemolytic transfusion reactions following emergency-release RBC transfusion. Am J Clin Pathol
18. St Clair CM, Shah M, Diver EJ, et al. Adherence to evidence-based guidelines for preoperative testing in women undergoing gynecologic surgery. Obstet Gynecol
19. Practice guidelines for obstetric anesthesia: an updated report by the American Society of Anesthesiologists Task Force on Obstetric Anesthesia and the Society for Obstetric Anesthesia and Perinatology. Anesthesiology
20. Van Klei WA, Moons KG, Rheineck Leyssius AT, et al. A reduction in type and screen
: preoperative prediction of RBC transfusions in surgery procedures with intermediate transfusion risks. Br J Anaesth
21. Fernández AM, Cronin J, Greenberg RS, et al. Pediatric preoperative blood ordering: when is a type and screen
or crossmatch really needed? Paediatr Anaesth
22. Butwick AJ, Goodnough LT. Transfusion and coagulation management in major obstetric hemorrhage. Curr Opin Anesthesiol
23. Goodnough LT, Daniels K, Wong AE, et al. How we treat: transfusion medicine support of obstetric services. Transfusion
24. Gutierrez MC, Goodnough LT, Druzin M, et al. Postpartum hemorrhage treated with a massive transfusion protocol at a tertiary obstetric center: a retrospective study. Int J Obstet Anesth
25. Hoeltge GA, Domen RE, Rybicki LA, et al. Multiple red cell transfusions and alloimmunization. Experience with 6996 antibodies detected in a total of 159,262 patients from 1985 to 1993. Arch Pathol Lab Med
26. Bauer MP, Wiersum-Osselton J, Schipperus M, et al. Clinical predictors of alloimmunization after red blood cell transfusion. Transfusion