Hysteroscopic surgery is a minimally invasive gynecological diagnosis and treatment technique. It is widely used in the clinic because it has the advantages of minimal trauma, quick recovery, and short hospital stay for postoperative recovery. Leiomyomas are the most common pelvic tumors, while submucosal fibroids are a common cause of abnormal bleeding and infertility [1]. OHIAS refers to a series of complications, such as pulmonary edema and heart failure, that result from fluid overload caused by large amounts of dilatant fluid entering the blood during hysteroscopic surgery. Due to the long duration, high perfusion pressure, and large perfusion flow of complex hysteroscopic surgery, a large amount of liquid is absorbed into the patient’s circulatory system through the intrauterine venous sinus and salpingot-peritoneum [2]. This leads to OHIAS, which in severe cases can directly endanger the patient’s life and safety. A 2020 expert consensus on anesthesia management in hysteroscopic diagnosis and treatment in China found the most common complications of hysteroscopic surgery were uterine perforation (0.12%), OHIAS caused by excess distention fluid (0.06%), intraoperative bleeding (0.03%), air embolism syndrome (0.03%), bladder or intestinal injury (0.02%), and endometritis (0.01%) [3]. The relatively high incidence of uterine perforation and OHIAS was noted with concern.

Currently, the prevention of OHIAS usually involves reducing uterine pressure, shortening the operation time, and reducing the amount of dilation fluid into the uterus [4]. Individual differences exist for the safe threshold of dilation fluid intake, with a threshold range of approximately 750 mL to 1000 mL for the electrolyte free solution depending on the patients’ body weight, cardiopulmonary function and other complications [5]. As for the normal saline, the safe threshold of fluid intake is 2500 mL [6]. Although many rules have been formulated and implemented for hysteroscopic surgery, the risks of pulmonary edema and heart failure still exist due to the increased circulating blood volume caused by large amounts of perfusion fluid entering the circulatory system during surgery [7]. Therefore, the monitoring of changes in lung fluid and preconditioning at the right time are crucial for reducing postoperative complications.

Ultrasonic monitoring of the lungs can accurately assess the condition of lung fluid. This is well established, performed routinely, and not limited by the time, place or cost of the procedure [8, 9, 10]. Lung B ultrasound conducted with an ultrasound probe shows imaging characteristics based on the different “air/water” ratios of lung tissue. Line A indicates the lung tissue is normal, whereas line B indicates the lungs are full of water. Thus, an increasing B line indicates gradually increasing pulmonary fluid [11]. During B-ultrasound examination, each side of the chest wall is divided into four zones, with a total of eight zones on both sides. The left side of the lung scan is from the second rib to the fourth rib, and the right side is from the second rib to the fifth rib. The sensor probe is scanned along the parasternal, mid-clavicle, anterior axilla and midaxillary line, thereby recording all B lines in the eight zones [12, 13]. The recognized positive index for B-ultrasound to indicate pulmonary edema is three or more B-lines in any two areas at the same time [10].

In the present study, B-ultrasound was used to monitor changes in lung water. Intravenous furosemide was administered when positive indicators of pulmonary edema appeared. The perioperative effects of pre- and post-treatment with furosemide were then compared. The aim was to determine the optimal time point for preconditioning in order to prevent pulmonary edema during hysteroscopic leiomyoma surgery. This could provide anesthesiologists with a strategy to prevent pulmonary edema during surgery, thus also reducing the probability of complications.

The doctors who conducted B-ultrasound monitoring in this study had received more than 6 months of professional training in pulmonary ultrasound monitoring.

This study was a randomized controlled clinical trial conducted in Fuxing Hospital, Capital Medical University, China. The study was approved by the institutional ethics committee (trial registration: 2022FXHEC-KY003), and all subjects signed informed consent. The study was registered at https://www.chictr.org.cn (registration number ChiCTR2200062840).

The inclusion criteria were: (1) age 20–60 years; (2) American Society of Anesthesiologists (ASA) grade 1–2; (3) patient scheduled for hysteroscopic surgery, with furosemide being used during surgery; (4) submucosal myoma (diameter $\ge$3 cm) complicated with bleeding, pain and other uncomfortable symptoms. The exclusion criteria were: (1) untreated hypertension; (2) condition affecting hemodynamics, such as heart failure; (3) thyroid disease; (4) adrenal disease; (5) emergency patient; (6) acute or chronic renal dysfunction (abnormal creatinine and urea nitrogen); (7) impaired liver function (alanine aminotransferase $>$200, or aspartate aminotransferase abnormality); (8) use of alpha-blockers and heart rate 55 beats/min; (9) body mass index (BMI) $>$30 kg/m${}^2$; (10) history of severe obstructive sleep apnea hypopnea syndrome (OSAHS) (apnea-hypopnea index $>$40); (11) history of alcoholism or recreational drug use; (12) history of allergy to drugs; (13) psychological or mental disorder; (14) autonomic nervous disease; or (15) diabetes.

A total of 62 patients who received hysteroscopic surgery in the Hysteroscopic Center of Fuxing Hospital, Capital Medical University from March 2022 to December 2022 were randomly divided into two groups using a computer-generated random number table. The numbers were sealed in envelopes and later opened by the main investigator to assign the allocated treatment. Eligible patients were allocated to the furosemide post-treatment group (n = 31) or to the furosemide pre-treatment group (n = 31). The process for assigning participants to each group is shown in a flow chart in Fig. 1. Patients and surgeons were blind to the assigned group.

Fig. 1.

Flow chart.

Group 1: 31 cases in the furosemide post-treatment group; 20 mg of intravenous furosemide when the lung auscultation was a wet rales sound.

Group 2: 31 cases in the furosemide pre-treatment group; the change in pulmonary water was monitored by ultrasound scanning during operation, and 20 mg of furosemide was injected intravenously when three B lines appeared simultaneously in any two zones.

During induction, the two groups received sufentanil (0.2 µg/kg) and propofol initiated at 6 µg/mL via target-controlled infusion (TCI). The two groups then received continuous propofol during the maintenance phase, with the dosage adjusted to maintain the bispectral index (BIS) at 45–55. The monitor used was an Infinity Vista XL (Dräger, Lubeck, Germany), and the anesthesia machine was a Fabius 2000 (Dräger, Lubeck, Germany).

All other procedures were the same in both groups. The end-expiratory carbon dioxide partial pressure was maintained at 35–45 mmHg during operation, with BIS maintained at 45–55.

Hysteroscopic leiomyoma surgery is a minimally invasive surgery. The uterine orifice was fully expanded and physiological salt was pumped into the uterine cavity at 70–80 mmHg pressure prior to the hysteroscopic operation. Pathological examination of the specimens confirmed they were uterine fibroids.

In both groups, patient data was recorded for age, body mass index, sum of the three diameters of the uterus, sum of the diameter of submucosal myoma, and intimal thickness. The mean arterial pressure and heart rate were recorded at four time points: entry into the operating room, after anesthesia induction, at the end of operation, and after laryngeal mask pulled out. Also recorded for each patient were the duration of anesthesia, recovery time after withdrawal of anesthetic drug, amount of sufentanil administered, amount of propofol administered, amount of uterine perfusion fluid, the amount of perfusion fluid flowing out, the actual amount of fluid intake (the difference between the amount of infusion and outflow), and the uterine perfusion pressure during operation. Postoperative adverse reactions (drowsiness, nausea, vomiting, chills), the proportion of pulmonary diffuse B-line in perioperative period (Intercostal space 50% line B) were also noted.

SPSS 26.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Continuous variables that conformed to a normal distribution (Kolmogorov-Smirnov test) were presented as the mean $\pm$ standard deviation and analyzed using the independent sample t-test. Other variables were presented as the median (interquartile ranges, IQR) and analyzed using the Wilcoxon test. Categorical variables were presented as frequencies (percentages) and analyzed using the chi-square test. Two-sided (except for the chi-square test) p-values 0.05 were considered statistically significant.

A total of 62 patients were enrolled in the study, with 31 cases in each group (Fig. 1). Surgery was successful in all patients. No statistically significant differences were observed between the two groups for patient age, body mass index, sum of the three diameters of the uterus, sum of the diameter of submucosal myoma, and intimal thickness. Furthermore, no significant differences between the two groups were seen for the mean arterial pressure and heart rate at the four time points at which these were assessed (Table 1).

No significant differences were observed between the two groups for the duration of anesthesia, amount of sufentanil, amount of propofol, amount of uterine perfusion fluid, the amount of perfusion fluid flowing out, the actual amount of fluid intake, or uterine perfusion pressure. However, there was a statistically significant difference in the recovery time between the two groups (p = 0.037) after withdrawal of the anesthetic drug (Tables 2,3).

There were no significant differences in the incidence of drowsiness, nausea or vomiting between the two groups. However, the incidence of postoperative chills was significantly lower (p = 0.032) in the pre-treatment group, and the proportion of pulmonary diffuse B-line in perioperative period was significantly lower also in the pretreatment group (p = 0.016) (Table 4).