Relationship Among Surgical Fibroid Removal, Blood Pressure, and Biomarkers of Renin–Angiotensin–Aldosterone System Activation

Abstract

Fibroids are common benign neoplasms in women and have recently been associated with cardiometabolic risk factors including hypertension. The objective of this study is to determine whether fibroid removal is associated with lower blood pressure (BP). We performed a single-center IRB-approved retrospective chart review of patients undergoing hysterectomy/myomectomy for fibroids and other benign gynecological surgical procedures from January 2016 to December 2019, and a prospective cohort study of patients undergoing hysterectomy/myomectomy for fibroids from August 2021 to April 2022. We measured demographic factors, preoperative/postoperative BP on day of surgery and at postoperative visit. In our prospective cohort, to evaluate for alterations in the renin–angiotensin–aldosterone system (RAAS) induced by fibroid removal, we measured serum angiotensin-II (Ang-II) and angiotensin converting enzyme (ACE) levels pre- and post-operatively. In our retrospective study (n = 294; mean age 41.9 ± 10.6, 43.5% Black, 50% with fibroids), we found that compared to patients without fibroids, patients with fibroids had significantly elevated systolic BP (SBP) (pre-op: p = 0.0005; post-op: p = 0.02), although this did not hold after adjusting for covariates. Fibroid removal was associated with a marginally albeit not statistically significant decrease in SBP (p = 0.062). In our prospective study (n = 11), there was a significant decrease in SBP following fibroid removal, but no change in diastolic BP (p = 0.019, p = 0.18, respectively). Serum levels of Ang-II and ACE were not significantly altered following surgical fibroid removal (p = 0.72, p = 0.81, respectively). Altogether, these findings suggest that fibroids are not independently associated with BP or RAAS activation, but do suggest that fibroid removal may be associated with a small drop in SBP.

Introduction

Uterine leiomyomas (fibroids) are monoclonal proliferations of uterine smooth muscle cells of the female reproductive tract. They are extremely common, affecting up to 70% of reproductive-age women and causing symptoms such as heavy or prolonged menstrual bleeding, pelvic pain, infertility, and bulk symptoms that negatively impact quality of life [1, 2]. Recently, a link has been drawn between fibroids and risk of cardiovascular disease, including hypertension [3–5]. It remains unknown, however, whether fibroids simply represent a physiological biomarker of hypertension and cardiovascular disease, or whether they may indeed contribute causally to cardiovascular disease.

Fibroids are intimately linked with the systemic vasculature through their feeding vessels, and they participate in a multitude of signal transduction circuits including endothelin-1, vascular endothelial growth factor (VEGF), and angiotensin pathways that may lead to vasoconstriction and endothelial dysfunction [6–8]. Other studies have identified genetic and epigenetic changes in fibroid tumors that provide crosstalk between the endocrine and immune systems, including translocation t(12;14) and 7q deletion as risk factors for fibroid recurrence, and DNA methylation and histone modification playing critical roles in fibroid pathogenesis [9]. Furthermore, polymorphisms in the angiotensin converting enzyme (ACE) and angiotensin receptor (ATR) genes have been associated with not only risk of hypertension but also risk of development of fibroids [10, 11]. These factors may explain the relationship between fibroids and hypertension [3]. Intriguingly, in a large nested case–control study of women with hypertension, those taking ACE inhibitors were significantly less likely to develop clinically significant fibroids over the course of 5 years [12].

To begin to answer the question of whether fibroids contribute causally to hypertension, it is important to know whether surgical treatment of fibroids has any influence on blood pressure (BP). We thus examined preoperative and postoperative BP among women undergoing surgery to remove fibroids (with a histologically confirmed diagnosis of fibroid on final pathology), along with demographically similar control women undergoing gynecologic surgery for benign indications.

The aim of this study was to establish whether fibroids are associated with elevated systolic or diastolic BP (SBP or DBP, respectively) and to determine whether surgical fibroid removal is associated with a decrease in either SBP or DBP. Additionally, we sought to determine whether fibroid removal leads to alterations in the renin–angiotensin–aldosterone system (RAAS).

Materials and Methods

We conducted a retrospective cohort study (January 2016–December 2019) using a sample of 313 women who underwent surgery to remove fibroids (cases) or other gynecological surgeries (controls) at a single academic medical center who consented to participate. This study was approved by our Institutional Review Board (IRB), Protocol # IRB00229605. Women were recruited into the biorepository of uterine fibroids, endometriosis, and adenomyosis study. Recruited subjects provided written informed consent to participate in this biorepository study, including access to biological specimens and medical information.

We abstracted SBP and DBP preoperatively on day of surgery (in the preoperative holding area) and postoperatively at each patient’s next BP measurement (typically the scheduled postoperative visit, but occasionally another healthcare encounter including primary care visit or other specialty provider). Charts were reviewed to ensure that patients had not been started on antihypertensives in the interim. There was no explicit protocol for measuring BP; however, standard practice at our institution is to perform BP measurement with an automated BP cuff with patient rested, in semi-recumbent position, with arm at the level of the heart. A single BP reading was obtained per patient for each data point included in the final analysis.

Inclusion criteria were women ≥ 18 years old, undergoing surgery for benign gynecological indications, who had follow-up BP readings between day of surgery and up to 12 months postoperatively. For all fibroid cases, only women with pathologically confirmed fibroids were included. As a control group, we included patients undergoing benign surgery for non-fibroid indications (e.g., diagnostic laparoscopy, laparoscopic endometriosis excision, laparoscopic cystectomy or salpingectomy, or hysterectomy for non-fibroid indication). In the fibroid group, we included women undergoing transvaginal (hysteroscopic), laparoscopic, and open myomectomies. Of 146 patients with fibroids undergoing myomectomy, there were 30 transvaginal myomectomies and 116 port site or abdominal myomectomies. Exclusion criteria were those with incomplete follow-up BP data, and those with postoperative BP measured either less than 2 days postoperatively or greater than 12 months postoperatively, and in the fibroid arm, those whose final pathology was negative for fibroids.

We also conducted an IRB-approved prospective cohort study (August 2021–April 2022), Protocol # IRB00196175 (cardiometabolic risk factors associated with uterine fibroids, myomectomy, and hysterectomy). Women provided written informed consent to participate in this study, including collection of blood samples and access to medical information. We conducted a power analysis to determine the number of subjects needed to detect a difference in serum biomarkers following fibroid removal. We predicted a 10% difference in serum levels of Ang-II and ACE, with a standard deviation of + / − 10% with 80% power, correlation of 0.5 between pre- and post- measures, and an alpha value of 0.05, which would require 10 subjects. We thus recruited a sample of 11 women undergoing hysterectomy or myomectomy for a preoperative diagnosis of uterine fibroids (which were all confirmed on final pathology and in whom we had complete pre and postoperative data from 9 subjects).

In our prospective cohort, we collected blood samples preoperatively, measuring serum levels of angiotensin-II (Ang-II) and angiotensin converting enzyme (ACE) via commercially available sandwich ELISA (The Johns Hopkins Core Laboratory Facility) from patients in the semirecumbent position in the pre-operative holding area, to be able to measure baseline RAAS activity in these patients. We then collected postoperative Ang-II and ACE at the time of the patients’ postoperative visit to determine whether fibroid removal was associated with altered RAAS signaling in vivo. Inclusion and exclusion criteria were identical to those in the retrospective study. We excluded patients with missing data.

Data from the retrospective cohort were analyzed using SAS v9.4 software (SAS Institute Inc, Cary, NC). Women with and without fibroids were compared using t tests or Wilcoxon rank sum tests for continuous variables and using Chi-squared or Fisher’s exact tests for categorical variables. Multivariable linear regression analyses predicted SBP and DBP pre- and post-operatively, as well as change in SBP and DBP from pre- to post-operation, adjusting for age, race, and presence of hypertension diagnosis. Data from the prospective cohort were analyzed with paired t tests and Wilcoxon rank-sum tests, and Fisher’s exact test, using SPSS^® software. A p value of < 0.05 was selected as the cutoff for statistical significance for all analyses.

Results

Of the 313 patients in the retrospective study, 19 were excluded due to age < 18 years (n = 1), no BP measurement (n = 1), no postoperative BP measurement (n = 5), and BP measurement only greater than 1 year after surgery (n = 12). There were 146 fibroid cases and 148 non-fibroid controls. Overall, for the n = 294 included patients, the mean age was 41.9 ± 10.6 years, 43.5% of subjects were Black, 42.1% were White, and the remainder fell into other race categories. Ninety one point five percent of patients were non-Hispanic. One hundred forty-six (49.7%) patients had fibroids, and 148 (50.3%) did not; 41% percent underwent hysterectomy. Comparing patients with and without fibroids, Black patients were significantly more likely to have fibroids, consistent with previous literature ([13], Table 1). There was no baseline difference in ethnicity (Hispanic versus non-Hispanic), presence of HTN or CAD among those with versus without fibroids (Table 1). Those with fibroids were significantly more likely to have undergone hysterectomy compared to those without fibroids (Table 1).

Table 1.

Differences in factors between patients with and without fibroids

Variable	Fibroid (n = 146)		No fibroid (n = 148)		P value
	N	Mean (SD) or Median [25th percentile-75th percentile]	N	Mean (SD) or Median [25th percentile-75th percentile]
Age at surgery (years)	146	43.84 (8.29)	148	40.06 (12.13)	0.002
EBL (mL)	144	50 [10–117.5]	140	10 [5–20]	< 0.0001
Length of stay in hospital (days)	146	0 [0–1]	147	0 [0–0]	< 0.0001
Fibroid weight (g)	126	255.15 [156.3–543]	0	NA	NA
BMI	146	33.18 (7.89)	143	31.91 (8.2)	0.181
Gravidity	145	2 [1–4]	148	2 [1–4]	0.623
Parity	145	2 [0–2]	148	2 [0–3]	0.973
Preoperative SBP	146	128.51 (14.36)	148	122.03 (17.17)	0.0005
Preop DBP	146	77.64 (10.5)	148	75.91 (10.93)	0.165
Postop SBP	146	126.31 (14.49)	148	122.36 (14.89)	0.022
Postop DBP	146	76.71 (11.6)	148	74.97 (11.06)	0.189
Days between surgery and BP	146	36 [25–45]	148	37 [20–50.5]	0.646
		N (%)		N (%)	P value
Race					0.001
Am Indian		1 (0.69)		1 (0.67)
Asian		6 (4.14)		8 (5.37)
Black		80 (55.17)		48 (32.21)
Other		10 (6.9)		15 (10.07)
Undisclosed		1 (0.69)		0 (0)
White		47 (32.41)		77 (51.68)
Ethnicity					0.232
Hispanic		9 (6.21)		14 (9.4)
Non-Hispanic		136 (93.79)		133 (89.26)
Undisclosed		0 (0)		2 (1.34)
Hypertension					0.103
Yes		60 (41.38)		48 (32.21)
No		85 (58.62)		101 (67.79)
Coronary artery disease					1
Yes		2 (1.38)		2 (1.34)
No		143 (98.62)		147 (98.66)
Hysterectomy					< 0.0001
Yes		97 (66.9)		25 (16.78)
No		48 (33.1)		124 (83.22)

Continuous variables with Gaussian distributions are presented as mean (SD) and were analyzed using t tests. Continuous variables with non-Gaussian distributions are presented as median [25th percentile–75th percentile] and were analyzed using Wilcoxon rank-sum tests. Categorical variables are presented as number (percent); categorical variables with a minimum of 5 in each cell were analyzed using Chi-squared tests, categorical variables with less than 5 in any cell were analyzed using Fisher’s exact tests

We next compared other patient-related factors and surgery-related factors between patients with and without fibroids. Patients with fibroids were slightly although statistically significantly older than those without fibroids (mean age 43.8 years vs. 40.1 years, respectively, p = 0.002, Table 1). Patients with fibroids also had a statistically significantly increased blood loss from surgery (mean 156.7 mL vs. 20.0 mL, respectively, p < 0.0001), and spent significantly more time in the hospital (0.53 vs. 0.12 days, p < 0.0001) (Table 1). BMI and parity were not significantly different between groups. Importantly, preoperative and postoperative SBP were significantly higher among patients with fibroids versus those without (p = 0.0005, p = 0.022, respectively), but there were no significant differences in postoperative DBP, and no difference in the number of days between BP measurements (Table 1). Mean time between BP measurements overall was 46.5 days (S.E.M. 2.9 days).

Comparing within groups, for patients with fibroids, we noted a non-statistically significant decrease in SBP pre- versus postoperatively of 2.2 mmHg (p = 0.062), and there was no difference in change in SBP for patients without fibroids (p = 0.138), and no changes in diastolic BP for those with or without fibroids (p = 0. 389, p = 0.999, respectively) (Table 2).

Table 2.

Change in systolic and diastolic blood pressure for patients with and without fibroids

Variable	Fibroid				No fibroid				Between group p value
	N	Mean (SD)	Range	Within group p value	N	Mean (SD)	Range	Within group p value
Change in SBP	146	− 2.21 (14.15)	− 35–32	0.06172	148	0.33 (15.07)	− 38–36	0.789655	0.138058
Change in DBP	146	− 0.93 (13)	− 30–41	0.387916	148	− 0.93 (12.16)	− 37–37	0.352392	0.999497

To determine whether fibroids would be independently associated with SBP or of DBP in our sample, we performed multiple linear regression analyses, controlling for age at surgery, Black race, and HTN. We found that when controlling for these co-variates, there was no significant independent effect of fibroids on SBP, DBP, or change in SBP from preoperative to postoperative time point (Table 3).

Table 3.

Adjusted multivariable linear regression results

Outcome modeled	Beta estimate for women with fibroids compared to no fibroids	Standard error of Beta estimate	95% Confidence Limits of Beta estimate		P value
Pre-operative SBP	2.6124	1.688	−0.696	5.9209	0.1217
Pre-operative DBP	0.9413	1.2561	−1.5207	3.4033	0.4536
Post-operative SBP	0.1668	1.5706	−2.9115	3.2451	0.9154
Post-operative DBP	0.6261	1.2814	−1.8855	3.1377	0.6251
Change in SBP from pre- to post-operation	− 2.4456	1.7672	−5.9093	1.018	0.1664
Change in DBP from pre- to post-operation	− 0.3152	1.5175	−3.2895	2.6591	0.8354

All models were adjusted for age at surgery, Black race (yes/no), and hypertension diagnosis (yes/no)

We next sought to determine whether patients with fibroids would have alterations in RAAS activation at baseline and following surgical fibroid removal. We recruited 11 participants with fibroids undergoing myomectomy or hysterectomy, of which 9 completed preoperative and postoperative lab work. We measured pre- and postoperative BP as well as serum levels of angiotensin-II (Ang-II) and angiotensin converting enzyme (ACE) preoperatively and at the time of the patients’ postoperative visit (mean time between measurements). To ensure internal validity, we confirmed fibroid diagnosis on final pathology.

Characteristics of the patients in this prospective cohort study are displayed in Table 4. We compared baseline Ang-II levels from the patients in our study (preoperative) to a historical control of women without fibroids from a previously published dataset [14]. We found no significant differences in Ang-II levels between these two groups (Fig. 1). Importantly, we found that systolic BP was significantly decreased following fibroid removal (change from preoperative to postoperative = 16 mmHg, two-tailed paired Student’s t test, p = 0.019), and a small but not statistically significant decrease in diastolic BP following fibroid removal (change from preoperative to postoperative = 7 mmHg, two-tailed paired Student’s t test, p = 0.18) (Fig. 2). We found no significant change in levels of Ang-II or ACE following fibroid removal (Ang-II change: − 0.5 ng/L preoperatively to postoperatively; ACE − 0.6 U/L preoperatively to postoperatively, two-tailed paired Student’s t tests, p = 0.76, p = 0.81, respectively) (Fig. 3, Table 4). These results suggest that fibroid removal is associated with a significant decrease in SBP but not associated with a significant change in RAAS activation.

Table 4.

Characteristics of patients in prospective cohort study (n = 11)

Characteristic	Median (range)
Age at surgery (years)	43 (32–51)
Parity	0 (0–2)
BMI (kg/m2)	31.6 (19.8–51.5)
Fibroid weight (g)	408 (99–1530)
EBL (mL)	50 (5–850)
Time between measures (days)	31.5 (8–53)
Pre-operative SBP (mmHg)	134 (110–163)
Pre-operative DBP (mmHg)	88 (58–99)
Post-operative SBP (mmHg)	118 (103–152)
Post-operative DBP (mmHg)	81 (68–93)
Change SBP (mmHg)	16
Change DBP (mmHg)	7
Pre-operative ANG-II (ng/L)	13 (0–74)
Post-operative ANG-II (ng/L)	12.5 (20–38)
Pre-operative ACE (U/L)	33 (18–106)
Post-operative ACE (U/L)	32.4 (20–77)
Change Ang-II (ng/L)	− 0.5
Change ACE (U/L)	− 0.6
	Proportion
Race
Black	9/11
White	1/11
Other	1/11
Ethnicity
Hispanic	1/11
Non-Hispanic	10/11
Hypertension	7/11
CAD	0/11

Fig. 1.

Comparison in baseline angiotensin-II levels between patients with fibroids (prospective cohort) and without fibroids (cohort of patients without fibroids derived from previously published historical control [14]). Two-tailed Student’s t test, p = 0.76

Fig. 2.

Comparison of systolic and diastolic blood pressure preoperatively versus postoperatively for patients in the prospective cohort study. Two-tailed Student’s t tests, p = 0.019, p = 0.18, respectively

Fig. 3.

Comparison of serum levels of ACE and Ang-II pre- versus postoperatively for patients in the prospective cohort study. Two-tailed Student’s t tests, p = 0.72, p = 0.74, respectively

Discussion

In this study, we set out to determine whether fibroids are associated with hypertension, and whether surgical fibroid removal would be associated with changes in BP and biomarkers of RAAS system activation. To our knowledge, this is the first study investigating the relationship between fibroid removal and change in BP. In our retrospective study, we observed significantly higher pre-operative and post-operative SBP in patients undergoing fibroid removal compared to those undergoing other benign gynecological procedures, before controlling for other covariates. We also noted a small, albeit not statistically significant decrease in SBP in patients undergoing fibroid removal. We found, consistent with prior literature, that fibroids were associated with hypertension and Black race [15]. However, after controlling for race and age, we did not observe a significant association between fibroids and systolic or diastolic BP. Others and we have reported associations between fibroids and cardiovascular disease, but a causal link has still yet to be conclusively drawn [3, 16]. The results of this study suggest that fibroids are not an independent risk factor for hypertension; however in our prospective study, we did observe a statistically and clinically significant decrease in SBP following fibroid removal of 16 mmHg. Given our small sample size, this result will need to be repeated to draw firm conclusions. Another question that this study did not address but should be investigated in the future will be the effect of pharmacological treatment of fibroids on BP. For instance, Linzagolix, a GnRH-antagonist currently under investigation for treatment of fibroids, suppresses the hypothalamic-pituitary-ovarian access and decreases heavy menstrual bleeding related to fibroids [17]. Whether benefits of this therapy extend to improved BP or other cardiovascular outcomes remains to be seen.

It is important to note that most myomectomies in our study were performed either laparoscopically, open, or in a minority of cases (20.5%) transvaginally (all hysteroscopically). Others have sought to determine whether route of fibroid removal may affect surgical outcomes. For instance, Lagana et al. [18] found that route of removal was not significantly associated with operative time, blood loss, complication rate, or hospital stay, although there was significantly less need for postoperative rescue pain medication with the transvaginal approach. Transvaginal colpotomy for fibroid extraction was not performed in the current study; however, this is a viable approach in certain circumstances with low complication rates [19]. Whether route of fibroid removal may influence blood pressure remains unknown, and it would be important to control for fibroid size in future work investigating this question (e.g., smaller fibroids may be more amenable to transvaginal extraction and may have a smaller impact on BP).

Systolic BP is most heavily influenced by arterial compliance, while diastolic BP depends both on total peripheral resistance and arterial compliance [20, 21]. Arterial compliance in patients with essential hypertension is generally decreased, and so it is possible that the fibroid uterus leads to reduced arterial compliance [22]. In this study, we observed a non-statistically significant change in ACE and Ang-II after fibroid removal. We hypothesized that fibroids participate in endocrine signaling through the renin–angiotensin–aldosterone system, increasing circulating levels of ACE and Ang-II to increase systemic vascular tone, thus leading to hypertension, and that fibroid removal would lead to a drop in these circulating factors. The current study does not support this mechanism.

The present study has several strengths and limitations which bear mentioning. In terms of strengths, the design included both a retrospective and prospective component. Secondly, we included both an internal control group (using each subject as her own control preoperatively to postoperatively) and an external control group (fibroid versus no fibroid). We also had a diverse group of subjects in our study, with 43% Black, 42% white, and 13% other races, potentially increasing the generalizability of our findings. Finally, unlike some studies which rely on patient-reported history of fibroids or imaging findings of fibroids, for each of our subjects with a diagnosis of fibroids, this was pathologically confirmed, increasing the precision of our results. In terms of limitations, our sample sizes for the retrospective and prospective components of the study were relatively small and do bear replication in larger cohorts of patients. We were underpowered for our primary endpoints of Ang-II and ACE levels, having complete data for only 9 out of the intended 11 subjects. We also relied on a single measurement of BP taken in a hospital or office setting. Ideally, we would have had a standardized BP measurement protocol (e.g., averaging three BP measurements after the subjects had been rested for at least 5 minutes, sitting upright in a chair with arm at heart level, resting comfortably with both feet planted on the floor), as is standard practice for measuring BP accurately. We also noted significantly different demographic and surgical characteristics between patients with fibroids (older, greater proportion Black, greater chance of having undergone hysterectomy) versus control patients without fibroids, potentially confounding our results. Future studies should aim to match these features to be able to draw direct comparisons more firmly. Notably, race is a sociological and not a biological construct, and the results of this study do not imply a biological basis for the link between Black race and fibroids.

In summary, we have shown that fibroids, while a biomarker of cardiovascular disease, do not independently contribute to elevated BP, although they may be associated with elevated SBP under certain conditions. Future work will be required to elucidate whether there are any potential benefits of fibroid removal on cardiovascular health.

Data Availability

All data and materials are available on request from the corresponding author.