Abstract
Aim: Our aim was to assess the clinical utility of postoperative hemoglobin testing following hysterectomy. Patients and Methods: We carried out a retrospective cohort study of patients who underwent robotic surgery at an academic center during a 44-month study period. Data included demographics and perioperative outcomes. The mean postoperative decrease in hemoglobin level was evaluated using numerical and categorical variables. Results: A total of 201 women were included. A total of 45 (22.4%) developed symptoms suggestive of hemodynamic compromise. When compared to asymptomatic patients, these patients were no different in operative time, estimated blood loss, pre- or post-operative hemoglobin levels, or the change in hemoglobin levels. Symptomatic patients did receive less fluid intraoperatively (1.2 vs. 1.5 l; p<0.0001). Perioperative outcomes were not associated with a greater postoperative decrease in hemoglobin (Hb). Postoperative anemia was associated with preoperative anemia (0% vs. 45%; p<0.0001). Patients with postoperative anemia were also more likely to be re-admitted within 30 days after surgery (7% vs. 23%; p=0.025). Of the three patients who received blood transfusions postoperatively, all three had preoperative Hb<9.5 g/dl, compared to 2.5% of those who were not transfused (p<0.0001). Using Institutional charges and Medicare reimbursement rates for blood hemoglobin testing, savings were estimated to be $3,629 and $1,236, respectively, during the study period. Conclusion: Postoperative Hb testing may be safely avoided unless starting Hb is less <10 g/dl. Clinical practice change can reduce healthcare costs without hindering patient care.
Uterine cancer is the most common gynecological cancer, with an estimated 65,950 new cases and 12,550 deaths in 2022 in the United States (1). The cornerstone of treatment for this malignancy is hysterectomy. Minimally invasive surgery has become the standard of care due to advantages over open surgery based on perioperative outcomes and survival (2, 3). The robotic platform (da Vinci® surgical System, Intuitive Surgical, Inc., Sunnyvale, CA, USA) was cleared by the Food and Drug Administration for gynecological surgery in 2005. In a randomized controlled trial that compared robotic-assisted surgery versus laparoscopy for endometrial cancer, the median estimated blood loss was 50 ml (4), which is significantly lower than under traditional open hysterectomy with an estimated blood loss of 291 ml in a systematic review (5). In that randomized study, the median decrease in hemoglobin on postoperative day 1 was 1.5 g/dl for both traditional laparoscopic and robotic-assisted surgery (4). The blood transfusion rate after robotic hysterectomy for endometrial cancer reportedly ranges from 0% to 5% (5).
The clinical utility of routine perioperative hemoglobin assessment has been evaluated in orthopedic, prostate, and benign gynecology surgery (6–11) but not in robotic-assisted hysterectomy for endometrial cancer. The cost associated with routine postoperative hemoglobin testing has been studied. A hospital-based study estimated annual potential savings of $2,804,662 if blood hemoglobin levels were not ordered after total laparoscopic hysterectomies for benign indications (9). Another study suggested cost savings of $1.96 million per year for pelvic organ prolapse repair after minimally invasive surgery (7).
It is not uncommon to routinely check the postoperative hemoglobin level after surgery for endometrial cancer regardless of the surgical approach to ensure there is no internal bleeding, and the patient can be safely discharged. This is probably based on historical practice rather than evidence-based guidelines. Given the overall low risk of complications and low estimated blood loss after robotic hysterectomy for endometrial cancer, our study’s aim was to assess the role of the postoperative hemoglobin level in patients with endometrial cancer who underwent robotic-assisted hysterectomy.
Patients and Methods
This was a retrospective cohort study in patients with newly diagnosed endometrial cancer who underwent robotic-assisted surgery from April 2016 to December 2019. The local Institutional Review Board approved the study (IRB201903205).
The primary objective of our study was to assess the clinical utility of postoperative hemoglobin testing after surgery. The secondary objective was to assess perioperative outcomes and cost.
Eligibility criteria. Adult patients with biopsy-proven endometrial cancer who presented to our academic center for scheduled robotic surgery were included. Patients with benign pathology, on chronic anticoagulation, having received preoperative blood transfusion, or who underwent open operation or conversion to open operation were excluded.
Data abstraction. Data were extracted from the patients’ electronic medical records in Epic (Epic Systems Corporation, Verona, WI, USA). Variables included: age, body mass index [BMI, calculated as weight (kg)/height (m2)], race, medical comorbidities, chronic anticoagulation, history of venous thromboembolism event, American Society of Anesthesiologists classification, prior surgery, operative time, estimated blood loss, intraoperative fluid volume, ketorolac use, uterine weight, tumor histology, International Federation of Obstetrics and Gynecology grade, stage, procedure, surgical complexity, intraoperative and postoperative complications, postoperative vital signs (heart rate, blood pressure), postoperative urine output, preoperative and postoperative hemoglobin level, postoperative transfusion, hospital length of stay, 30-day re-admission, and repeat hemoglobin testing during the current admission.
Robotic-assisted surgery followed the surgical principles described in the National Comprehensive Cancer Network’s guidelines (12). Perioperative care was performed according to Enhanced Recovery After Surgery protocols (13, 14). It is our institutional practice to admit surgical patients overnight with tentative discharge planned for within 23 hours from surgery.
Data analysis. Categorical variables are presented as the number of cases and percentages. Continuous variables are presented as median, mean, standard deviation, and range. The chi-square or Fisher exact test was used as appropriate to compare categorical variables, and t-tests were used to compare continuous normally distributed variables. Pearson’s correlation coefficient was used to correlate perioperative variables with the change in hemoglobin level on postoperative day 1. Statistical analysis was performed with SPSS v.22 (IBM Corp, Armonk, NY, USA). Potential annual cost savings were estimated using institutional and Medicare reimbursement rates for complete blood count testing and estimated new endometrial cancer cases.
Results
A total of 201 patients met the inclusion criteria. Two patients were excluded: One received blood transfusion during surgery due to severe anemia and coronary artery disease, and the other received blood transfusion during her admission but before surgery due to severe anemia (Hb=6 g/dl).
The mean age of the cohort was 63.1 years with a mean BMI of 36.4 kg/m2. 71.6% were White, 16.9% were Black, and 6.5% were Hispanic. The majority of patients (60.2%) had had prior abdominal surgery. An American Society of Anesthesiologists score of 3 was recorded for 80.6% of the patients. Demographics and clinical characteristics of the cohort are presented in Table I.
Patient demographics.
The median estimated blood loss was 50 ml and median intraoperative intravenous fluid was 1500 ml. All patients were admitted for a period of at least 23-h observation after their procedure. All patients had their hemoglobin checked prior to surgery and on postoperative day 1. The median preoperative hemoglobin level was 13.2 g/dl, with 10 patients having an initial hemoglobin level of less than 10 g/dl. The median postoperative hemoglobin level was 11.7 g/dl, with 22 patients having level of less than 10 g/dl. The mean postoperative decrease in hemoglobin level was 1.37 g/dl. The median length of hospital stay was one day. Twenty-five patients required repeat blood work due to either being symptomatic or having unexpected abnormal results. Three patients (1.5%) required blood transfusion. Surgical and perioperative outcomes are outlined in Table II. In 57.7% of cases, surgery was started before 12:00 pm, and 173 (86%) underwent hysterectomy, bilateral salpingo-oophorectomy, and sentinel lymph node mapping or node dissection.
Surgical and perioperative outcomes of study patients (n=201).
Forty-five patients (22.4%) developed symptoms and signs suggestive of hemodynamic compromise (at least one episode of tachycardia and one episode of hypotension). When compared to asymptomatic patients, these patients were no different in operative time, estimated blood loss, pre- or post-operative hemoglobin levels, or change in hemoglobin level (p>0.05). Symptomatic patients did, however, receive less fluid intraoperatively (1.2 l vs. 1.5 l; p<0.0001) (Table III).
Comparison of hemoglobin (Hb)-related parameters between patients with and without postoperative clinical evidence of potential anemia (n=201)*.
When compared to patients with postoperative hemoglobin >10 g/dl, patients with postoperative anemia (hemoglobin <10 g/dl, 22 patients) were no different in demographics (American Society of Anesthesiologists score, age, race, BMI), intra-operative variables (uterine weight, operative time, estimated blood loss, intraoperative intravenous fluids), or postoperative outcomes (tachycardia, hypotension, oliguria, or changes in hemoglobin level). The only perioperative parameter associated with postoperative anemia (Hb <10 g/dl) was preoperative anemia (0% vs. 45% for these groups, respectively; p<0.0001). Moreover, these patients with postoperative anemia were more likely to be re-admitted within 30 days after surgery (23% vs. 7%; p=0.025) (Table IV).
Comparison of parameters between patients with and without postoperative anemia (Hb ≤10 g/dl).
Of the three patients who received blood transfusions postoperatively, all three had preoperative hemoglobin level of <9.5 g/dl, whereas among the patients who did not receive a blood transfusion, only 2.5% had an initial hemoglobin level of <9.5 g/dl (p<0.0001) (Table V). Blood transfusion was not correlated with demographics (American Society of Anesthesiologists score, age, race, BMI, comorbid conditions, surgical history, cancer stage and grade), operative outcomes (surgery complexity, uterine weight, operative time, estimated blood loss, intraoperative intravenous fluids), or postoperative outcomes (tachycardia, hypotension, oliguria, or change in hemoglobin level).
Comparison of hemoglobin (Hb) levels between patients with and without postoperative blood transfusion.
There was also no correlation between perioperative variables and the mean change in the postoperative hemoglobin level (Table VI).
Correlation between perioperative variables and mean change in postoperative hemoglobin level.
In 2021, the institutional cost for a complete blood count and the Medicare reimbursement rate for blood hemoglobin testing were $19 and $6.47, respectively. During the study period, potential savings amounted to $3,629 and $1,236, respectively.
Discussion
We evaluated the clinical utility of postoperative hemoglobin after robotic-assisted surgery for endometrial cancer. Our findings suggest that routine hemoglobin testing has little clinical benefit following robotic hysterectomy and should be reserved for patients with a preoperative hemoglobin of less than 10 g/dl. Substantial cost savings would result from judicious use of hemoglobin level testing following robotic hysterectomy. Clinical practice change can reduce healthcare costs without hindering patient care.
We recommend judicious use of postoperative hemoglobin level testing following robotic hysterectomy and to consider checking the postoperative hemoglobin level for patients with anemia preoperatively. Patients with preoperative iron deficiency anemia should be medically optimized before surgery with oral iron without delaying cancer treatment. A mathematical model to predict postoperative day 1 hematocrit levels after hysterectomy was described by Swenson et al. (15). In a validation study, the equation correctly predicted hematocrit to within ±5% for 83.2% of women. The equations included preoperative hematocrit, weight, estimated blood loss, intraoperative fluid, and preoperative platelet count. However, the study population also included ovarian cancer, cervical cancer, and all surgical approaches (laparoscopy, abdominal, and robotic surgery). With an estimated blood loss of 167.5 ml, the equation may not be fit for most patients who undergo robotic surgery endometrial cancer (16).
We identified that symptomatic patients (22.4%) received less fluids intraoperatively (1.2 l vs. 1.5 l; p<0.0001). It is possible that their symptoms were due to dehydration at the time of surgery since almost half of the cases were performed after 12:00 pm (42.3%). Adequate hydration during and after surgery may help avoid unnecessary measures, such as repeat blood testing or more active monitoring.
To our knowledge, this is the first study to evaluate the clinical utility of postoperative hemoglobin testing after surgery for endometrial cancer. The unique characteristics of this operation include the exploration of the pelvic sidewall for lymph nodes and comorbid conditions in this particular patient population. Exploration of the nodes along the aorta, and common, external and internal iliac vessels may increase the risk of bleeding. Two papers were published in the literature on benign gynecology assessing the role of postoperative hemoglobin testing after minimally invasive surgery. Chamsy et al. included 629 patients who underwent total laparoscopic hysterectomies for benign reasons. Overall, 2.5% developed symptoms or signs suggestive of hemodynamic compromise. When compared to asymptomatic patients, symptomatic patients had a larger postoperative decrease in hemoglobin level (2.66 vs. 1.8 g/dl) and were more likely to undergo blood transfusion, pelvic imaging or re-operation. Factors associated with a larger postoperative decrease in hemoglobin level were a low BMI and higher intraoperative intravenous fluid volume. They concluded hemoglobin testing had little clinical benefit and recommend testing only for patients who develop signs or symptoms of acute anemia (9). Lavelle et al. included patients who underwent robotic (326, 40.7%) and laparoscopic surgery (474, 59.3%) for sacrocolpopexy. The mean postoperative decreased in hemoglobin was 1.76 g/dl. More than half of their patients had clinical evidence of potential anemia, but few (5%) had a postoperative hemoglobin level ≤10 g/dl and none required blood transfusion. Women with clinical evidence of potential anemia had a lower postoperative hemoglobin level (11.57 vs. 12.19; p<0.001) and larger mean hemoglobin decrease (1.91 vs. 1.49; p<0.001). The only factor associated with larger hemoglobin decrease was lower BMI. They concluded that postoperative hemoglobin testing was rarely beneficial to clinical care (7).
Given the low estimated blood loss and low risk of postoperative complications, there are proponents of same-day discharge after robotic-assisted hysterectomy for gynecological malignancy (17, 18).
Routine postoperative testing has financial implications and should be constantly re-evaluated. Bulger et al. identified five opportunities for improved healthcare value. One of their recommendations was to avoid repetitive complete blood count and chemistry testing in the face of clinical and laboratory stability (19). Published data suggest that for most patients with endometrial cancer, the estimated blood loss is low, and the risk of transfusion is 0-5% (5).
It is estimated that in 2022 there will be 65,950 new cases of uterine cancer in the United States (1). Assuming 90% of these cases will undergo minimally invasive surgery and 96% will have a preoperative hemoglobin level greater than 9.5 g/dl (extrapolating from our results), a total of 56,981 cases will have minimally invasive surgery with an initial hemoglobin level greater than 9.5 g/dl. Using institutional charges and Medicare reimbursement rates for blood hemoglobin testing ($19 and $6.47, respectively), omitting testing would lead to estimated cost savings of $1,082,639 and $368,667, respectively.
The strengths of our study include the use of consecutive cases undertaken by the same surgical team with experience in robotic surgery using a standardized approach, which produced consistent, precise, and reliable data. The limitations of our results are inherent to study design and potential selection bias. However, as far as we are aware, no other published data exist assessing the role of postoperative hemoglobin testing in patients who underwent robotic-assisted surgery for endometrial cancer.
Conclusion
Hemoglobin level testing has little clinical benefit following elective robotic hysterectomy and should be reserved for patients with preoperative anemia. Substantial cost savings would result from the judicious use of postoperative hemoglobin level testing following robotic hysterectomy.
Acknowledgements
Abstract was presented as a poster at the 2021 American College of Surgeons Quality and Safety Conference on July 12-16, 2021.
Footnotes
Authors’ Contributions
Massoud Shoraka, M.D.: Conceptualization, data collection, investigation, methodology, writing – review and editing. Semiramis L. Carbajal-Mamani, M.D.: Conceptualization; data curation; investigation; methodology; project administration; resources; visualization; writing – original draft, review and editing. Ali Zarrinpar, M.D., Ph.D.: Data curation, formal analysis, investigation, methodology, supervision, validation, visualization, writing – review and editing. Curtis Warren, M.P.H., C.P.H.: Data curation, formal analysis, investigation, supervision, writing – review and editing. Bahram Dideban, M.D., M.Sc.: Conceptualization, data collection investigation, methodology, supervision, validation, writing – review and editing. Bernie Amaro, B.A.: Investigation, methodology, project administration, resources, writing – review and editing. Joel Cardenas-Goicoechea, M.D., M.B.A.: Conceptualization; data curation; investigation; methodology; project administration; resources; visualization; writing – original draft, review and editing.
Conflicts of Interest
The Authors report no conflicts of interest in relation to this study.
- Received February 4, 2022.
- Revision received February 22, 2022.
- Accepted February 25, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.





