Comprehensive assessment of the postoperative period after uterine artery embolization based on stress response markers and Doppler imaging of uterine blood flow reduction

Nowadays, uterine fibroids are one of the commonest and most challenging problems encountered in gynecological practice. Globally, it is estimated to affect 20% of women over 30 years and more than 40% of women over 40 years [1]. According to the SOGC Clinical practice guideline (2015), uterine fibroids are the commonest gynecological tumors, with a prevalence of 70% to 80% in women who have reached the age of 50.

Uterine artery embolization (UAE) is a safe, effective, organ-sparing, and less invasive alternative to surgery widely used for the treatment of uterine fibroids [3-6]. In the immediate postoperative period after UAE, more than 50% of patients develop a postembolization syndrome, which typically occurs after embolization of any solid organ and is thought to be an immune-mediated response [7, 8].

As with any surgical injury, in response to UAE, the body initiates a systemic response comprising an endocrine and systemic inflammatory (immune) stress response [9]. To adequately protect the body, it is necessary to control stress, hemodynamic, endocrine and metabolic reactions, prevent and/or reduce pain and the accompanying emotional reactions [9–12]. Studying stress markers, controlling systemic reactions’ manifestations in minimally invasive surgical procedures, such as UAE, alleviating of the postembolization syndrome and the optimizing anesthesia remains a challenging research problem.

The study aimed to comprehensively assess the postoperative period after uterine artery embolization (UAE) based on the dynamics of clinical and immuno-biochemical indicators of the stress response and Doppler imaging of uterine blood flow reduction; to investigate the relationship between the intensity of the immuno-biochemical response, pain syndrome severity, and ultrasound-detected arterial blood flow reduction in fibroids. Another aim was to identify the features of the systemic inflammatory response in patients undergoing UAE.

Material and methods

This pilot study comprised 60 patients with symptomatic multiple uterine fibroids, who underwent UAE at the Department of Gynecology of the RICEL Clinic. Inclusion criteria were as follows: verified diagnosis of multiple interstitial or subserosal uterine fibroids, ASA (American Society of Anesthesiologists) physical status class I to II, and the absence of pain caused by concurrent diseases.

Before UAE, all patients underwent laboratory testing and clinical evaluation, including hysteroscopy and fractional diagnostic curettage with the subsequent histologic investigation, and pelvic ultrasound examination. The ultrasound protocol included transabdominal and transvaginal echography by the standard technique, transabdominal and transvaginal color Doppler and Doppler color flow mapping (DCFM). Doppler imaging was used to observe peripheral and internal fibroid arteries and uterine arteries. The analysis of blood flow characteristics included the maximum possible number of visualized color voxels, which determined the nature of blood flow (arterial or venous), systolic and diastolic blood flow velocity, and resistance index (RI) – the ratio of (peak systolic velocity – end diastolic velocity) to peak systolic velocity.

UAE was performed in interventional radiology operating room. Vascular access was obtained under local anesthesia with lidocaine through the right femoral artery by the standard Seldinger technique. Contour spherical PVA micro-particles with diameters ranging from 710 to1000 microns and MeriMedical 500-700 micro-particles were used as embolic agents. Iopamidol 370 mg/ml was used as a contrast agent. The mean operative time was 43 ± 0.9 min. Angiographic criteria for the effectiveness of the UAE were as follows: 1) uterine artery occlusion (stasis of contrast in the uterine artery); 2) marked opalescence of fibroids; 3) retrograde reflux of the contrast agent from the uterine artery; 4) early arteriovenous reflux.

Anesthesia was done using both opioid and non-opioid analgesics according to standard indications to ensure adequate analgesia. As a non-opioid, centrally acting analgesic we used nefopam. It was given intravenously in single doses of 20 mg before UAE; at the time of embolization, patients were given an intravenous injection of 20 mg of promedol.

Postoperative analgesia in all groups consisted of non-opioid analgesics (nefopam 20 mg 6–8 hours ± ketoprofen 50 mg every 6–8 hours) and opioid analgesics according to indications.

Effectiveness was assessed based on changes in hemodynamic parameters and levels of stress response markers. In the postoperative period, patients were asked to fill in a pain diary and rate their pain intensity using the visual analog scale (VAS) and assess changes in their perception of pain. If pain was present, the patients were asked to evaluate the intensity of pain and its effect on daily activities, sleep and the quality of life.

Analysis of the immuno-biochemical stress reaction profile included serum levels of pro-inflammatory (interleukin (IL) -1, IL-6, tumor necrosis factor α (TNF α)) and anti-inflammatory (IL-4, IL-10) cytokines, C-reactive protein, hormones (stress hormones): adrenocorticotropic hormone (ACTH), cortisol; levels of glucose and lactate.

Serum cytokines measurement was performed using an enzyme immunoassay test systems: Human IL-1 beta Platinum ELISA Bender MedSystems GmbH, Austria BMS224HS; IL-10 Platinum ELISA, BMS215 / 2 BMS215HS, Bender MedSystems GmbH, Austria; Human TNF-alpha Platinum ELISA, BMS223 / 4; BMS223HS Bender MedSystems GmbH, Austria; CRP, HS (C-Reactive Protein) EIA3954 DRG International Inc., USA. ACTH concentrations were determined in EDTA-treated plasma using an IMMULITE 2000 analyzer, Siemens (Germany), and serum cortisol was measured in an IMMULITE 2000 analyzer, Siemens (Germany). The glucose level in plasma was determined by the hexokinase method using a Beckman Coulter AU480 analyzer (Japan); C-reactive protein was determined using immunoturbidimetric assay, and plasma lactate was determined by an enzymatic method.

The hemodynamic profile assessment was done using the Nihon Kohden monitor (Japan) and included ECG, heart rate, non-invasive blood pressure measurements (systolic, diastolic and mean arterial pressure), and hemoglobin oxygen saturation.

The control of the above parameters was carried out in the following steps: 1 – upon the patient arrival to the operating room (baseline); 2 – 2 hours after occlusion of the uterine arteries; 3 – 24 hours after surgery; 4 – 48 hours after surgery.

Pelvic ultrasound was performed on a Voluson-E8 Expert BT-12 system using multi-frequency transabdominal (2.5–5 MHz) and transvaginal (3.6–8 MHz) transducers. Ultrasound examination was performed before surgery and on the third day after UAE.

The study was approved by the local ethics committee. Written informed consent was obtained from all patients enrolled in the study.

Statistical analysis was performed using Statistica Version 7.0. The normality of the distribution was tested by the Shapiro-Wilk test. For variables showing normal distribution, the results were expressed as the mean (M) and the standard error of the mean (m), and the data were presented as a mean ± standard error of the mean (M ± m). Non-normally distributed variables were expressed as the median (Me), the first (Q1) and the third (Q3) quartiles, and the data were presented as the median and interquartile range (Me (Q1; Q3). Qualitative variables were expressed as numbers and percentages. To determine the statistical significance of the differences between the variables obtained at different points in time, we used the Friedman test. To compare the results with baseline data, the paired Student’s ?????-test and Wilcoxon test were used for normally and non-normally distributed variables, respectively. Differences between the groups were considered statistically significant at p<0.05. The correlation was estimated as strong (r> 0.75), moderate (0.25 <r <0.75) and weak (r <0.25). For correlation, the significance level was set at p <0.05.

Results and discussion

The mean age of patients was 43.3 ± 1.19 years (from 31 to 56 years); body mass index was 26.58 ± 1.06. All patients were classified as ASA PS class I and II, with a predominance of ASA PS class II (76.6%). Comorbidities were categorized by a stage of compensation or remission.

The fibroids’ diameter did not exceed 76.7 mm (mean 39.5 ± 1.68 mm). All study participants had multiple (from 2 to 8) fibroids (mean 4.75 ± 1.87). Fibroids had different locations (patients with subserosal pedunculated fibroids were excluded).

Occlusion of the uterine arteries and reduction of blood supply to fibroids triggers a stress-response, clinically manifested by a postembolization syndrome and changes in metabolic, endocrine and immune parameters.

In the immediate postoperative period, no significant changes were observed in hemodynamics. Mean arterial pressure did not exceed 80.9 ± 1.86 mm Hg. without a significant increase from the baseline values ​​at all stages of observation. On days 1- 3, 21 patients (35%) developed subfebrile fever, which went up to a maximum of 37.7 °C in two patients. At the same time, an increase in leukocyte counts to 12.44 ± 1.87 thousand/μl was noted.

The development of pain is a sensory and emotional reflection of alteration. Pain syndrome of varying degrees of intensity was observed in 100% of patients. Painful sensations began to increase 4–8 hours after UAE and maintained at a maximum level during 1-2 days postoperatively (Table 1). The pain was accompanied by psychological discomfort and sleep disorders. Eighteen (30% of 60) patients had quite severe pain ranging 7 through 10 scores on VAS, which they would not like to experience again.

In recent years, a significant interdependence of the neuroendocrine and mediator cytokine systems has been established. The rise in pro-inflammatory cytokines stimulates secretion of ACTH, corticosteroids, and catecholamines, and acute-phase proteins, that is, increases the intensity of the endocrine stress response. In turn, activation of the hormonal stress response stimulates the production of cytokines [10, 12]. Of course, the increase in the studied immuno-biochemical parameters was not as intense as in the case of extensive surgical interventions. However, an increase in the concentrations of ACTH, cortisol, and glucose in our study reaches statistically significant values ​​compared with the baseline values, as well as an increase in serum levels of the inflammatory response markers. The observed changes in the humoral and immune status were unidirectional. Two hours after surgery, the levels of both cytokines and stress hormones began to increase, reaching a maximum after 24–48 hours.

Assessment of the endocrine-metabolic component of the stress response showed fluctuations in the levels of ACTH and cortisol occurring within reference ranges for these variables. Nevertheless, there was a statistically significant increase in their concentrations compared with baseline levels with maximum levels of ​​24–48 h after UAE (p <0.05). The maximum values ​​of ACTH were recorded after 24 hours and amounted to 15.17 ± 0.53 pg/ml (194% from a baseline of 7.81 ± 0.36 pg/ml), p <0.05. At the same time, serum concentration of cortisol reached 13.8 ± 0.78 μg/dl, i.e.158% of the baseline 8.72 ± 0.64 μg/dl (p <0.05).

Surgical aggression results in stress-induced hyperglycemia due to excessive glycogenolysis in the liver, caused by sympathetic adrenergic stimulation. The stress-induced hyperglycemia is proportional to the intensity of surgical trauma. Two hours after the UAE, the glycemic level was statistically significantly higher than the baseline values.

The concentration of lactate did not rise beyond the reference range; no statistically significant changes were observed during the study. The changes in endocrine and metabolic variables are given in Table 2.

The production of cytokines increases in proportion to the severity of surgical aggression and reflects the invasiveness of surgical intervention [10, 13]. After UAE, we observed an increase in the concentrations of both pro-inflammatory (IL-1, IL-6, TNFα) and anti-inflammatory cytokines (IL-4, IL-10) with statistically significant differences with their baseline levels ​​(p <0.05). The peak level of pro-inflammatory cytokines coincided in time with that of and anti-inflammatory cytokines that were recorded on the second or third day after UAE. The level of IL-1 increased from baseline and amounted to 6.84 ± 0.11 pg/ml (167% of the baseline 4.10 ± 0.14 pg/ml), p <0.05. More significant changes were noted postoperatively in the levels of IL-6, which increased from 4.30 ± 0.13 pg/ml at baseline to 9.38 ± 0.26 pg/ml (218% of the initial level), p <0.05. TNFα concentration increased from 0.53 ± 0.04 pg/ml at baseline to 0.84 ± 0.03 pg/ml (158% of the baseline level), p <0.05.

Concentrations of pro-inflammatory cytokines were increasing in parallel with that of anti-inflammatory cytokines. Thus, 48 hours after UAE, the concentration of IL-10 increased from 5.34 ± 0.44 pg/ml to 7.90 ± 0.78 pg/ml (147% of the baseline level), p <0.05. Similar changes were observed in the levels of IL-4, which increased from 2.21 ± 0.13 pg/ml to 3.49 ± 0.11 pg/ml (158% of the baseline level), p <0.05.

There was a significant increase in the level of C-reactive protein, which is a sensitive indicator of tissue damage. In the early postoperative period, its concentration increased from 1.57 (1.17; 3.96) to 9.45 (6.95; 11.48) mg/l (602% of the baseline level), p <0.05. Changes in cytokines and C-reactive protein are shown in Table 3.

The ultrasound criterion for the effectiveness of UAE was considered to be a complete loss of fibroid arterial perfusion after UAE. On the 3rd day after UAE, 56 patients (93.3% out of 60) had no intra-fibroid blood flow, three patients (5% out of 60) had peripheral blood flow through the capsule of one of the fibroids; the remaining fibroids were avascular. Thus, in 98.3% of the UAE was effective; only in one patient (1.7% of 60) DCFM visualized voxels of intra-fibroid blood flow, which was associated with the anatomical features of the uterine blood supply in this patient, namely, the preservation of the active blood supply to the fibroid from the ovarian artery after complete obstruction of the uterine arteries.

Besides, on the 3rd day after the UAE, DCFM showed a marked reduction of uterine artery blood flow with a decrease in the peak systolic velocity from 51.9 ± 5.6 cm/sec to 15.1 ± 5.1 cm/sec (p <0.05) and an increase in the RI from 0.69 ± 0.09 to 0.78 ± 0.04 (p <0.05).

In arcuate arteries outside the zones of nodular deformity, the peak systolic velocity decreased from 30.2 ± 0.5 cm/sec to 12.7 ± 2.4 cm/s (p <0.05), RI increased from 0.64 ± 0 11 to 0.74 ± 0.09 (p <0.05). These findings are consistent with the literature [14–16]. On the third day after the UAE, ultrasound examination demonstrated that the size of the uterus and fibroids not only did not decrease but on the contrary, increased. The volume of the uterus was 207.05 ± 11.8 cm3 (p <0.05), with baseline averages of 186.66 ± 7.4 cm3; the mean fibroid volume increased from 39.5 ± 1.68 mm3 at baseline to 43.42 ± 2.16 mm3. These changes seem to be associated with the development of aseptic necrosis in the fibroids, perifocal edema and inflammation of the uterine wall in the postembolization period.

Blood flow reduction in fibroids and impaired uterine perfusion are the pathophysiological basis for the development of the postembolization syndrome, as evidenced by the strong correlation between these parameters and the level of the increase in stress markers (IL-1, IL-6, TNFα, IL-4, IL-10, C-reactive protein, ACTH, cortisol, glycemia). For all monitored indicators, the correlation coefficient exceeded 0.75 (p <0.05). Correlation analysis showed a moderately strong (r = 0.68) correlation between pain intensity and the degree of uterine blood flow reduction (p <0.05). Analysis of the relationship between the degree of increase in immuno-biochemical parameters and the patients’ subjective perception of pain showed a moderately strong correlation between the pain intensity and the increase in IL-1 and IL-6 (r=0.48, р<0.05 and r=0.56, р<0.05, respectively). In other cases, the correlation coefficient at all stages of the study did not exceed 0.18–0.25 (p <0.05), which corresponds to a weak correlation.

Conclusion

In the postembolization period, there was a statistically significant increase in the serum concentrations of pro-inflammatory (IL-1, IL-6, TNFα) and anti-inflammatory (IL-4, IL-10) cytokines, C-reactive protein, levels of stress hormones (ACTH, cortisol), and glycemia. The intensity of the stress response depends on the extent of the fibroid blood flow reduction and uterine arterial perfusion after UAE with a strong correlation between these variables.

Changes in the immuno-biochemical profile develop in parallel with the intensity of pain syndrome and ultrasound signs of aseptic necrosis of fibroids and perifocal inflammation in the uterine wall. Patients’ subjective perception of pain has a moderate correlation with the changes in immuno-biochemical parameters. Also, a moderately strong correlation was found between the pain scores on VAS and the decrease in uterine perfusion, which once again testifies to the versatility of the neurophysiological processes of pain and psycho-emotional factors in pain perception.

Pain syndrome, as one of the main manifestations of the postembolization syndrome, was observed in 100% of patients. About 1/3 of the patients had quite severe pain ranging 7 through 10 scores on VAS, which they would not like to experience again. Therefore, further optimization of anesthesia and postoperative analgesia when performing UAE is warranted.