Umbilical Morcellation and Postoperative Pain in Patients Undergoing Benign Hysterectomy

Abstract

Background:

Morcellation has allowed patients with enlarged uteri to obtain a minimally invasive hysterectomy with improved outcomes; however, there is little information regarding postoperative pain. Our study aims to compare pain scores and opioid requirements in patients undergoing umbilical morcellation during benign minimally invasive hysterectomy versus those who do not require morcellation.

Methods:

A retrospective cohort study was performed at a tertiary care center including patients who underwent total laparoscopic or supracervical hysterectomy by one high volume surgeon from 2019 to 2022. Baseline characteristics, postoperative pain scores, and morphine milligram equivalents in the acute and late setting were recorded. Two-sample t test for continuous variables and χ² or Fisher’s exact test for categorical variables were used to compare differences. A multiple regression model evaluated the effect of groups with the adjustment of confounders.

Results:

A total of 232 patients underwent hysterectomy in which 57 underwent umbilical manual morcellation and 175 did not. There was no difference in postoperative complications, readmissions, or blood products required (P > 0.05). Individuals that underwent umbilical morcellation had longer operative times (226.6 vs 120.1 minutes, P < 0.01), more blood loss (311.1 vs 82.0 mL, P < 0.01), longer length of stay (0.60 vs 0.44 days, P < 0.01), increased uterine weight (1,293.2 vs 151.6 g, P < 0.01), and fibroid pathology (93.0% vs 46.3%, P < 0.01). There were no differences in postoperative pain scores, immediate and later opioid use between groups on multivariate analysis.

Conclusions:

Patients who undergo umbilical morcellation, typically for large fibroid uteri, have similar postoperative pain scores, opioid use, and postoperative complications to those who undergo hysterectomy for other indications.

INTRODUCTION

Minimally invasive hysterectomy has improved patient outcomes compared to abdominal hysterectomy by lowering postoperative pain scores, decreasing time to return to normal activities, and decreasing complication rates.^1–3 Morcellation has allowed patients with enlarged uteri, most commonly due to fibroids, to obtain a minimally invasive hysterectomy with improved outcomes.^2,4,5 Given concern regarding malignant and benign tissue dissemination with a power morcellator,^6–10 other interventions have been employed to decrease tissue spillage including in-bag morcellation and manual morcellation techniques.^11–15

Tissue extraction can be performed either by mini laparotomy, typically from extending the umbilical incision, or vaginally through the colpotomy.^16,17 Data demonstrates there is no difference in perioperative outcomes regardless of umbilical or vaginal morcellation.^18,19 However, there is limited data regarding pain differences in patients who undergo morcellation versus those who do not. Two small prospective studies of a total of less than 60 patients showed there were no differences in pain scores in patients who either underwent vaginal or abdominal morcellation.²⁰ These studies did not evaluate pain in comparison to those who did not undergo any type of morcellation. Our goal was to evaluate postoperative pain in benign minimally invasive hysterectomy patients who undergo manual umbilical morcellation versus those who do not undergo morcellation.

MATERIALS AND METHODS

A retrospective cohort study was performed at an academic tertiary care center including patients who underwent a benign total laparoscopic or supracervical hysterectomy by one high volume surgeon from March 2019 to December 2022. IRB approval was obtained prior to collection of patient information. All patients underwent a minimally invasive hysterectomy with three to four 5 mm abdominal incisions. For patients that underwent umbilical morcellation after specimen detachment and placement in a containment bag (if possible due to size), the umbilical incision was extended to 3 cm and the specimen morcellated manually. Patients were excluded if they were on chronic opioid therapy, underwent suprapubic morcellation, or underwent an unrelated concomitant procedure with another service. Baseline patient characteristics were identified including race, age, American Society of Anesthesiologists (ASA) class, body mass index (BMI), presence of diabetes, smoking status, parity, prior abdominal surgeries, and indication(s) for surgery. Patients were identified as having a regional block if they had transversus abdominis plane or erector spinae plane blocks by anesthesia. Pathology reports were used to indicate final pathology as well as uterine weight. Immediate postoperative patient reported pain scores were transcribed from the medical record. Estimated blood loss (EBL), surgical time, length of stay (LOS), and postoperatively defined hernia were noted. Morphine milligram equivalents (MMeq) were calculated in the immediate perioperative period in the operating room and the post anesthesia care unit prior to discharge or admission. Late postoperative MMEq were calculated based on a prescription monitoring program and opioid use on the floor during their inpatient admission, if applicable. Patients that did not undergo morcellation were compared with those who underwent umbilical manual morcellation. Two-sample t test for continuous variables and the χ² or Fisher’s exact test for categorical variables were used to compare differences. Multiple regression was employed to evaluate the associations of covariates selected based on the statistical and clinical standpoints with 1-hour, 2-hour, and max pain scores as well as immediate and late MMeq. These covariates included age, race, morcellation status, BMI, ASA class, surgical time, EBL, LOS, regional block, uterine weight, chronic pelvic pain indication (cyclical and/or noncyclical), and fibroid indication.

RESULTS

A total of 263 patients who underwent a benign hysterectomy were identified. Two patients were excluded for suprapubic morcellation, 17 for vaginal morcellation, 6 for chronic opioid use prior to their procedure, and 6 for having a combination procedure. A total of 175 patients did not undergo morcellation while 57 underwent umbilical manual morcellation during their procedure. Of the umbilical morcellation group, 12 had supracervical hysterectomies while 45 had total hysterectomies. There were no differences in ASA class, BMI, diabetes status, smoking status, parity, prior vaginal deliveries and abdominal surgeries, or indication for surgery (P > .05). Patients in the umbilical morcellation group were older (44.3 vs 40.8 years, P < .01), more often Black (47.4% vs 26.3%, P < .01), and more likely to have fibroids as the indication for surgery (87.7% vs 20.6%, P < .01) (Table 1).

Table 1.

Baseline Characteristics

Characteristics	No Morcellation (N = 175) N (%)	Umbilical Morcellation (N = 57) N (%)	P Value
Age, mean	40.8 (10.5)	44.3 (7.0)	.004
Race
Black or African American	46 (26.3%)	27 (47.4%)	.005
Hispanic or Latino/Asian	16 (9.1%)	7 (12.3%)
White	113 (64.6%)	23 (40.3%)
ASA class
1	14 (8.0%)	4 (7.0%)	.969
2	95 (54.3%)	31 (54.4%)
3 or 4	66 (37.7%)	22 (38.6%)
Body mass index, mean	33.3 (9.3)	35.6 (11.4)	.138
Diabetes
Yes	19 (10.9%)	9 (15.8%)	.321
No	156 (89.11%)	48 (84.2%)
Smoking status
Former	14 (8.0%)	1 (1.8%)	.171
Never	133 (76.0%)	49 (86.0%)
Current	28 (16.0%)	7 (12.2%)
Prior abdominal surgery
Yes	120 (68.6%)	32 (56.1%)	.086
No	55 (31.4%)	25 (43.9%)
Parity, mean	1.50 (1.5)	1.3 (1.4)	.247
SVD
Yes	64 (36.6%)	21 (36.8%)	.971
No	111 (63.4%)	36 (63.2%)
Pain indication
Yes	74 (42.3%)	17 (29.8%)	.094
No	101 (57.7%)	40 (70.2%)
AUB indication
Yes	101 (57.7%)	29 (50.9%)	.366
No	74 (42.3%)	28 (49.1%)
Fibroids indication
Yes	36 (20.6%)	50 (87.7%)	<.001
No	139 (79.4%)	7 (12.3%)

Overall base characteristics such as age, race, and ASA in comparison with patients who underwent umbilical morcellation versus those who did not.

There was no difference in postoperative complications, readmissions, blood products required, or adenomyosis or endometriosis pathology identified between the groups (P > .338). There were no intraoperative visceral or vascular injuries in either group. There was one postoperative hernia identified in each group. Postoperative complications in the nonmorcellation group included a vaginal cuff dehiscence, one vaginal cuff superficial separation, three pelvic abscesses, one vaginal cuff cellulitis, one temporary brachial plexus paresthesia, and one enterocolitis. Postoperative complications in the morcellation group included one vaginal cuff dehiscence and one small bowel obstruction. Patients who underwent umbilical morcellation had a longer operative time (226.6 vs 120.1 minutes, P < .01), more blood loss (311.1 vs 82.0 mL, P < .01), longer LOS (0.60 vs 0.44 days, P < .01), increased uterine weight (1,293.2 vs 151.6 g, P < .01), and fibroid pathology (93.0% vs 46.3%, P < .01) (Table 2).

Table 2.

Surgical Outcomes

Characteristics	No Morcellation (N = 175) N (%)	Umbilical Morcellation (N = 57) N (%)	P Value
Operative time in minutes	120.1 (43.8)	226.6 (89.3)	<.001
Postop complications
Yes	8 (4.6%)	2 (3.5%)	>.999
No	167 (95.4%)	55 (96.5%)
Readmission
Yes	5 (2.9%)	0	.338
No	170 (97.1%)	57 (100.0%)
Blood products given
Yes	1 (0.6%)	1 (1.8%)	.401
No	174 (99.4%)	56 (98.3%)
Estimated blood loss, mean	82.0 (96.0)	311.1 (406.9)	<.001
Length of stay, mean	0.44 (0.32)	0.60 (0.35)	.002
Regional block
Yes	94 (53.7%)	30 (52.6%)	.887
No	81 (46.3%)	27 (47.4%)
Uterine weight, mean	151.6 (104.4)	1293.2 (1303.1)	<.001
Fibroids pathology
Yes	81 (46.3%)	53 (93.0%)	<.001
No	94 (53.7%)	4 (7.0%)
Adenomyosis pathology
Yes	49 (28.0%)	16 (28.1%)	.992
No	126 (72.0%)	41 (71.9%)
Endometriosis pathology
Yes	19 (10.9%)	4 (7.0%)	.400
No	156 (89.1%)	53 (93%)

Overall surgical outcomes such as operative time and postop complications compared with patients who underwent umbilical morcellation versus those who did not.

There were no differences in 1-hour, 2-hour, and max pain scores or immediate or late opioid use between the two groups upon basic comparison and multilinear regression (P > .05) (Table 3). On multiple regression, there were no differences in any pain scores or immediate or late opioid use based on age, surgical time, EBL, or uterine weight (P > .05). However, race, BMI, regional block use, and fibroid indication were noted to have differences in pain scores or opioid equivalents. Race was associated with different 1-hour pain scores. Black patients had an average of 0.91 and 1.44 greater scores than White or Hispanic/Latino or Asian individuals, respectively (P = .03). Race, BMI, regional blocks, and fibroid indication were associated with different 2-hour pain scores. Again, Black patients had an average of 1.12 and 2.04 greater scores than White or Hispanic/Latino or Asian individuals (P < .01). With one unit of increased BMI, the 2-hour pain score was increased by 0.05 (P < .01). Patients who had a regional block had an average of 0.98 less pain scores than those who did not. Patients with fibroids had lower pain scores than other patients (–1.20, P = .01). Race, ASA class, and LOS were associated with max pain scores. Black patients had an average of 1.8 greater scores than Hispanic/Latino or Asian individuals (P = .01). Those that had an ASA class of 3 or 4 had 0.95 greater pain scores than those that were an ASA class of 2. The max pain scores increased by 1.3 with each day increase in LOS. Immediate MMeq differences were associated with race and LOS. Black patients had 9.6 more MMEq than Hispanic/Latino or Asian patients (P = .03). There were no other differences with race. With every one day increase in LOS, the immediate MMeq were increased by 9.7 (P = .01). LOS and having an indication of pelvic pain correlated with more MMeq. With every one day increase in LOS, the immediate MMeq were increased by 32 (P < .01). Patients with chronic pelvic pain prior to their procedure required 13.4 (P = .04) more MMeq than those who did not.

Table 3.

Pain and Opioid Scores

Characteristics	Total Population (N = 148) N (%)	No Morcellation (N = 175) N (%)	Umbilical Morcellation (N = 57) N (%)	P Value
MMEq immediate, mean	38.5 (18.1)	37.6 (19.2)	41.12 (14.1)	.138
MMEq late, mean	103.8 (47.4)	102.6 (47.9)	107.6 (46.0)	.489
Opioid pills late, meana	13.8 (6.3)	13.7 (6.4)	14.3 (6.1)	.489
Pain, mean
1 hour (n = 232)	3.59 (2.62)	3.71 (2.64)	3.23 (2.53)	.224
2 hours (n = 218)	3.11 (2.60)	3.10 (2.63)	3.13 (2.54)	.943
Max pain score w/in 12 hours (n = 232)	5.37 (2.94)	5.46 (2.91)	5.11 (3.03)	.435

Multilinear regression comparison of hourly pain scores, immediate postoperative opioid use, late opioid use after the immediate postoperative period in all patients, those who underwent umbilical morcellation and those who did not. MMeq, morphine milliequivalents.

^aIn 5-mg oxycodone equivalents.

DISCUSSION

There were no differences in immediate postoperative pain, immediate perioperative opioid requirements, and late opioid prescriptions between patients requiring umbilical manual morcellation and those not requiring morcellation. Our study results are consistent with a prior study finding no differences in postoperative pain or opioid use in total laparoscopic versus supracervical hysterectomies.²¹ However, this study is limited to umbilical morcellation with only supracervical hysterectomies that do not require an additional vaginal incision unlike our study which included patients who underwent total hysterectomies with umbilical morcellation.

There were no differences in postoperative complications or readmissions between groups. This is consistent with prior research regarding perioperative complications with those undergoing umbilical morcellation versus those who did not, specifically supracervical versus laparoscopic assisted vaginal hysterectomies.¹⁹ Umbilical morcellation was associated with an increase in surgical time (100 minutes), EBL (222 mL), and uterine weight (1,093 g). These findings are likely secondary to the increased complexity of uterine fibroids. This is also consistent with literature showing a positive correlation with operating times and EBL based on uterine weight.²² Despite this variation, there was no difference in transfusion or postoperative complications in our study population. Our study population also had very large uterine weights (average 1,293 g, median 921 g in the umbilical morcellation group). Thus, our results further promote evidence that patients undergoing minimally invasive hysterectomy for large fibroid uteri requiring morcellation have similar complication rates than individuals undergoing a laparoscopic hysterectomy for other benign conditions.^22,23 Our findings also demonstrate that LOS less than 24 hours and same day discharge for patients undergoing surgery for large fibroids is safe and effective with 73.7% of patients going home the same day in the morcellation group (LOS 0.60 days, standard deviation [SD] 0.35 vs 0.44 days, SD 0.32 in the no morcellation group). Finally, we provide evidence that these patients do not have differences in immediate postoperative pain scores or opioid requirements than their nonmorcellation counterparts despite a larger incision.

The observed increase in MMeq consumption in individuals with chronic pelvic pain indication was expected as prior studies have shown patients with chronic pelvic pain, including dysmenorrhea, have increased rates of opioid consumption after hysterectomy.^24,25 It is important to note that patients between groups had no statistically significant difference in chronic pelvic pain indication prior to surgery as well as no pathology confirmed differences in endometriosis. These findings are significant as a history of endometriosis has been shown to increase postoperative opioid consumption.²⁴ Lastly, as expected, there was also increased opioid consumption noted in those who were admitted for observation for inadequate pain control.

Hourly pain scores were mixed between those in comparison to race, BMI, and indication. However, there were no major differences in MMEq in relation to these except LOS. Therefore, although there were minor differences in hourly scores depending on each factor, ultimately there were no differences in opioid administration. These changes in hourly pain scores could be secondary to time of administration of opioids as exactly hourly time periods may not correlate well with administration time. Additionally, there were a variety of patients that did or did not receive a bolus of opioids just prior to leaving the operating room, thus, opioids administered in the operating room was included and may contribute to this score difference.

Our study shows that regional blocks did not positively or negatively affect postoperative pain scores in our patients. The benefits of regional blocks on pain scores and/or opioid use for those undergoing minimally invasive hysterectomy have been mixed in the literature.^26,27

Strengths of this study include a larger surgical population than previously published. Additionally, one surgeon performed all hysterectomies, therefore, eliminating variation in surgical techniques and potential pain differences due to variations between surgeons. Furthermore, there were no major differences in surgical populations between groups including those undergoing hysterectomy with chronic pelvic pain.

Limitations of this study include patient reported postoperative pain scores as these were not a visual analog score, which our institution has not adopted. Second, although the number of opioids prescribed including refills was identified, the number of pills used was not able to be verified and may not reflect the exact use of medication, therefore, limiting the accuracy of late opioid use results. Lastly, this study cannot be translated into general practice as these results are from one experienced minimally invasive surgeon.

CONCLUSIONS

In this study, we found that there were no differences in postoperative complications, immediate postoperative pain and immediate and late postoperative opioid use between patients requiring umbilical morcellation and those not requiring morcellation at the time of laparoscopic hysterectomy. Given the very large uterine size of our morcellation population, it provides further evidence that these patients may safely undergoing minimally invasive hysterectomy with similar outcomes to those that undergo hysterectomy for other benign conditions in the hands of a skilled surgeon.