DOI: https://doi.org/10.21203/rs.3.rs-2131419/v1
The purpose of this study is to assess the impact of different temperatures and equilibrium times on the clinical outcomes of FET cycles during the thawing procedure and to select a better thawing method to improve clinical outcomes.
This retrospective study included 1,734 FET cycles from January 1, 2020, to January 30, 2022. Embryos vitrified using a Kitazato Vitrification Kit were thawed at 37°C in all steps (the case group, denoted the “all-37°C” group) or at 37°C and then at room temperature (RT; the control group, denoted the “37℃-RT” group), according to the kit instructions. The groups were matched 1:1 to avoid confounding.
After case-control matching, 366 all-37°C cycles and 366 37℃-RT cycles were included. The baseline characteristics were similar (all P > 0.05) between the two groups after matching. FET of the all-37°C group yielded a higher clinical pregnancy rate (CPR; P = 0.009) and implantation rate (IR; P = 0.019) than FET of the 37℃-RT group. For blastocyst transfers, the CPR (P = 0.019) and IR (P = 0.025) were significantly higher in the all-37°C group than in the 37℃-RT group. For cleavage embryo transfers, the CPR and IR were non-significantly higher in the all-37°C group than in the 37℃-RT group (P > 0.05).
Thawing vitrified embryos at 37°C in all steps during FET cycles can enhance CPR and IR. Well-designed prospective studies are warranted to further evaluate the efficacy and safety of the all-37°C thawing method.
The freezing and thawing of oocytes/embryos/blastocysts is an important step in assisted human reproduction procedures, the widespread use of which has increased the chances of pregnancy for women undergoing in vitro fertilization (IVF)–embryo transfer [1]. In particular, the use of frozen-thawed embryo transfer (FET) has increased dramatically in recent years, as it offers a high live-birth rate (LBR) and a low risk of adverse pregnancy outcomes such as preterm birth, low birth weight, small for gestational age, multiple pregnancies, and ovarian hyperstimulation syndrome, thereby enhancing the safety and efficacy of the treatment [2–6].
Embryos contain a high volume of water, the freezing of which must be regulated during cryopreservation, as ice crystals formed inside cells during cooling or warming may cause serious damage [7, 8]. Cryoprotectants are used to reduce physical damage during chilling, but they contain a variety of chemicals that in high concentrations can have toxic effects on and cause osmotic damage to cells [9, 10]. The two methods commonly used for preserving embryos/blastocysts/oocytes are slow freezing and vitrification [11]. Slow freezing uses a low concentration of cryoprotectant and a sufficiently low entire cooling rate to effectively dehydrate cells and reduce ice crystal formation [12]. In contrast, vitrification uses a high concentration of cryoprotectant and rapid cooling to vitrify the inside and outside of cells without producing ice crystals, and the cryoprotectant concentration inside the cell during vitrification is lower than that during slow freezing [13, 14], which reduces damage to cells during freezing. Compared with slow freezing, the use of vitrification for freezing embryos has less of an effect on cell metabolism and the genome and increases the survival and developmental potential of embryos [15, 16]. Vitrification has also been shown to improve clinical pregnancy outcomes compared with slow freezing; specifically, vitrification results in a higher LBR and does not increase unfavorable neonatal outcomes. Therefore, most fertility clinics are opting for vitrification over slow freezing [17–19].
Despite significant advances in the vitrification technique for assisted reproduction, a variety of parameters—such as cryoprotectant composition and concentration, equilibration duration, chilling, and warming temperatures and rates—damage cells and affect embryo survival and development [20–22]. The majority of studies have concentrated on the effects of freezing on embryo injury, survival, and development, but some have shown that the rate of thawing has a greater effect on embryos [8, 23–26]. That is, even though few or small ice crystals may have been formed during the freezing process, small ice crystals can recrystallize into larger ice crystals during the thawing process and cause embryo cell damage. For example, Bo Jin et al. found that embryos did not exhibit visible ice-crystal formation during freezing but were more likely to show visible ice crystals after slow thawing at moderate temperatures. Consequently, the survival rate of vitrified frozen embryos thawed at a moderate rate (25°C in the air for 120 s) was much lower than that of embryos that were thawed rapidly [7]. A rapid thawing rate prevents the recrystallization of small intracellular ice crystals and decreases cell damage, thereby increasing the survival and developmental potential of oocytes and embryos [27–30].
The Kitazato Vitrification Freezing and Thawing Kit (Nakajima, Fuji, Shizuoka, Japan) is one of the most commonly used vitrification kits in assisted human reproduction procedures. The freezing solution in this kit contains high concentrations of sucrose, ethylene glycol, and dimethyl sulfoxide. The thawing procedure suggested by the manufacturer involves two temperature steps: warming at 37°C followed by equilibrating at room temperature (RT) (i.e., incubating embryos in thawing solution [TS] at 37°C for 1 min, in dilution solution [DS] at RT for 3 min, in washing solution 1 [WS1] at RT for 5 min, and WS2 at RT for 5 min). This method thaws oocytes or embryos mainly at RT to decrease the chemical toxicity of cryoprotectants. Truong et al. investigated the effect of antioxidants on frozen-thawed blastocysts using the Rapid-Warm Blast Kit (Vitrolife Sweden AB, V. Frӧlunda, Sweden). All thawing steps were performed at 37°C and excellent results were obtained [31]. However, it remains unknown which embryo thawing procedure is optimal for FET: thawing at 37°C in all steps (all-37°C) or at 37°C followed by RT (37°C-RT). To this end, we compared the FET outcomes of all-37°C embryos with those of 37°C-RT embryos. Our findings suggested that thawing at 37°C in all steps improves the clinical pregnancy rate (CPR) and implantation rate (IR).
Since April 1, 2020, the abovementioned protocols (all-37°C and 37°C-RT) for thawing embryos have been used at our center, but their relative clinical outcomes are unknown. Accordingly, we conducted a retrospective case-control matched study that included 1,734 FET cycles performed from January 1, 2020, to January 30, 2022, at the Department of Reproductive Genetics, the First Affiliated Hospital of Kunming Medical University. In this study, we evaluated the effects of the two thawing protocols on the recovery rate of embryos frozen by vitrification and the clinical outcomes after FET. Only women aged 38 or younger who had undergone conventional IVF or intracytoplasmic sperm microinjection (ICSI) insemination and were on their first FET cycle were included. Women were excluded if they had been diagnosed with recurrent miscarriage, uterine anomalies, or adenomyosis. We also excluded cycles with a thin endometrium (< 7 mm) and thawed oocytes, as well as no high-quality embryos or blastocysts (i.e., Grade III or IV cleavage stage embryos and blastocysts inferior to grade 3BB).
The study was conducted according to the Declaration of Helsinki and approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University, and no informed consent was required because the study was a retrospective study.
To enhance the reliability of this study, the required sample was determined based on the CPR for FET cycles performed from January 1, 2020, to January 1, 2021. The CPR was 56.2% per FET cycle during this period, from which it was calculated that 334 cycles would be needed in each group to achieve a power of 85% and detect a 20% between-group difference. Case and control groups were matched at a ratio of 1:1, reducing the interferences of common confounders.
The insemination methods of all FET cycles in our center were conventional IVF or ICSI. Oocytes after injection or zygotes 16–18 h after insemination were transferred separately to 30-µL of G-1™ PLUS and covered with mineral oil (Vitrolife Sweden AB, V.Frӧlunda, Sweden). Fertilization was observed under an inverted phase contrast microscope (Nikon, Japan) at a magnification of 200–400 ×. Embryos showing normal fertilization, defined as the presence of two clear pronuclei (2PN). Then, the dishes were placed in a 37 ℃ Multi-Gas Incubator (Astec, Japan, 7.0% CO2; 5.0% O2; 88% N2) to culture embryos. On Day 3 (D3), cleavage embryos were composed of 6–12 symmetric blastomeres, with < 20% fragmentation, normal zona pellucida, and no multinucleation was considered good quality [32, 33]. One or two cleavage embryos of good quality with the highest scores were used for embryo transfer or vitrification. The remaining usable cleavage embryos were transferred separately to 30-µL of G-2™ PLUS and covered with mineral oil (Vitrolife Sweden AB, V.Frӧlunda, Sweden) for blastocyst culture. On Day 5 or Day 6 (D5 or D6), the morphology of blastocysts was evaluated according to the Gardner&Schoolcraft scoring system. Expanded blastocysts were evaluated as stages 3–6 based on the degree of expansion of the blastocyst cavity; the appearance of inner cell mass and trophectoderm was scored as A, B, or C [34]. One or two blastocysts with a grade of 3BB or higher were selected for embryo transfer or vitrification.
Kitazato Vitrification Kit VT101 (Nakajima, Fuji, Shizuoka, Japan) was used for vitrification, and all steps in the vitrification procedure were performed at RT according to the kit instructions. First, the equilibrium solution (ES; 7.5% v/v ethylene glycol; 7.5% v/v dimethyl sulfoxide; 20% v/v replace and supplement serum; 0.005% v/v gentamicin; 83.995% v/v 199HEPES buffer solution) and the vitrification solution (VS; 17% v/v sucrose; 15% v/v ethylene glycol; 15% v/v dimethyl sulfoxide; 20% v/v replace and supplement serum; 0.005% v/v gentamicin; 51.995% v/v 199HEPES buffer solution) were stood at RT (20–25°C) for at least 30 min. The selected embryos were placed in ES for 8 min and then washed by transfer via pipette to three successive microdroplets of VS by pipetting. After the final wash, the embryos were immediately placed in a Cryotop (Kitazato Corporation), then the Cryotop was quickly immersed in liquid N2 (-196°C). The period from the transfer of the embryo into the first VS microdroplet to the immersion of the embryo into liquid N2 was equal to 45–60 s. Before freezing, blastocysts were manually shrunk with a laser to allow fluid from the blastocyst cavity to flow out [35].
The two protocols employed for thawing cryopreserved embryos with the Kitazato Thawing Kit VT102 (Yanagishima, Fuji, Shizuoka, Japan) were as follows. In the 37°C-RT protocol, the TS (34% v/v sucrose; 20% v/v replace and supplement serum; 0.005% v/v gentamicin; 64.995% v/v 199HEPES buffer solution) droplets were pre-warmed in a 37°C Multi-Gas Incubator without gas (Astec, Japan, 7.0% CO2; 5.0% O2; 88% N2) for at least 30 min. Then the DS (17% v/v sucrose; 20% v/v replace and supplement serum; 0.005% v/v gentamicin; 81.995% v/v 199HEPES buffer solution) and WS1 and WS2 (20% v/v replace and supplement serum; 0.005% v/v gentamicin; 98.995% v/v 199HEPES buffer solution) were aspirated and covered with mineral oil at 37°C to rewarm before being exposed to RT, which could reduce the osmotic pressure change of solutions. The vitrification-preserved embryos were placed in the TS droplet at 37°C for 1 min, then placed in a DS droplet for 3 min, then transferred to a WS1 droplet for 5 min, and finally transferred to a WS2 droplet for 5 min. All steps of embryo movement into DS to WS2 were performed on a workstation at RT. In the all-37°C protocol, the TS, DS, WS1, and WS2 were pre-warmed for at least 30 min at 37°C under the same conditions. All subsequent thawing steps were performed on a 37°C workstation. The embryos were kept in the TS droplet for 1 min, then transferred to the DS droplet for 3 min, then to the WS1 solution for 2.5 min, and finally to the WS2 droplet for 2.5 min. The blastocyst-thawing procedure was the same as the embryo thawing procedure. Finally, the embryos were promptly transferred to BM dishes and observed under an inverted phase contrast microscope (Nikon, Japan) at a magnification of 200–400 × to evaluate their development and morphology after thawing. Embryos that survived thawing were transferred to double-well plates and incubated for 2–4 h at 37°C in a Multi-Gas Incubator (Astec, Japan, 7.0% CO2; 5.0% O2; 88% N2).
Depending on the physiology of the patients, their endometria for FET were prepared following a natural cycle protocol, an ovulation-stimulation protocol, a hormone-replacement protocol, or a pituitary downregulation hormone-replacement protocol. When the thickness of the endometrium was greater than or equal to 8 mm, embryos were then transferred to the prepared endometrium. Blood was collected 12–14 days after the endometrial implantation to track serum hCG concentrations in real-time. Clinical pregnancy was defined as a positive β-hCG test (β-hCG > 10 mIU/mL), followed by ultrasound detection of a fetal heart or gestational sac 2–3 weeks later. An early miscarriage was defined as the ultrasound-detected disappearance of the gestational sac before the twelfth week of pregnancy.
The demographic characteristics and clinical indicators associated with FET cycles for both male and female partners were compared between the all-37°C (case) and 37°C-RT (control) groups. Characteristics and indicators compared were the age and body mass index (BMI) of the male partners and the age, BMI, type of infertility, serum anti-Müllerian hormone (AMH) concentration, endometrial preparation protocol, endometrial thickness, insemination method, number of oocytes aspirated, number of embryos thawed, and embryo transfer stage (D3/D5/D6) of the female partners.
The primary outcome was CPR. Secondary outcomes were the survival rate, early miscarriage rate (eMR), and IR. The CPR was defined as the total number of cycles with at least one gestational sac with positive cardiac activity at the seventh week of amenorrhea divided by the total number of FET cycles. The survival rate was calculated as the number of embryos with at least 50% of their cells surviving after thawing divided by the total number of thawed embryos. The eMR was defined as the total number of pregnancy losses before 12 weeks of gestation divided by the total number of transferred embryos. The IR was defined as the number of gestational sacs monitored by ultrasound divided by the number of transferred embryos.
The case and control groups were matched 1:1 for potential confounding factors, namely, the woman’s age, the embryo transfer stage (D3/D5/D6), the number of transferred embryos, and the endometrial preparation method.
The baseline characteristics of the two groups were compared. Continuous data were analyzed for normality using the Shapiro–Wilk test and normal distribution plots. As no continuous variables had a normal distribution at baseline, their values are presented as medians with interquartile ranges. Categorical variables are presented as numbers with percentages. Continuous variables were compared using the Mann–Whitney U test. The distributions of categorical variables were compared using a chi-square test or Fisher’s exact test. All tests were two-tailed, with P < 0.05 considered to be statistically significant. SPSS version 26 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis.
Figure 1 represents the patient selection process. 1,734 FET cycles met our inclusion criteria, 305 of which were excluded for the following reasons: female partner diagnosed with recurrent miscarriage, uterine abnormality, or adenomyosis (n = 91); no high-quality embryos or blastocysts (n = 169); endometrial thickness less than 7 mm (n = 32); and cycles with thawed oocytes (n = 13). Of the remaining 1,429 FET cycles, embryos in 369 cycles were thawed following the all-37°C protocol, and embryos in 1,060 cycles were thawed following the 37°C-RT protocol. After case-control matching, 366 cycles from each group were included in our analysis.
Table 1 summarizes the baseline characteristics of the two patient groups. There were no significant differences (P > 0.05) between the two groups in terms of the age or BMI of either partner. The female partners in each group were not significantly different (P > 0.05) in terms of the type of infertility, serum AMH concentrations, number of oocytes obtained, or insemination method. The female partners’ endometrial preparation protocols during the FET cycle were the same for both groups (P = 1), and there was no significant difference in the number of frozen embryos thawed (P = 0.930) or the embryo transfer stage (P = 0.332) between the groups.
Variables |
37℃-RT |
All-37℃ |
P-value |
---|---|---|---|
No. of cycles |
366 |
366 |
|
Female age, years, median (IQR) |
31(28–33) |
31(28–33.25) |
0.965 |
Male age, years, median (IQR) |
32 (29–35) |
32(29–35) |
0.955 |
Female BMI, kg/m2, median (IQR) |
21.8(19.8–24.2) |
22.35(20–24.6) |
0.151 |
Male BMI, kg/m2, median (IQR) |
23.7(21.3–26.3) |
24.05(22–26.4) |
0.180 |
Type of infertility, % (n) |
|||
Primary |
51.1(187) |
57.4(210) |
0.088 |
Secondary |
48.9(179) |
42.6(156) |
|
AMH, ng/mL, median (IQR) |
4.45(2.85–7.14) |
4.79(2.75–7.91) |
0.233 |
Endometrium preparation protocol, % (n) |
|||
HRT |
48.1(176) |
48.1(176) |
1 |
GnRHa-HRT |
49.7(182) |
49.7(182) |
|
Natural cycle |
1.6(6) |
1.6(6) |
|
Gn |
0.5(2) |
0.5(2) |
|
Endometrium thickness, mm, median (IQR) |
10(8–11) |
10(9–11) |
0.205 |
Fertilization, % (n) |
|||
IVF |
80.6(295) |
82(300) |
0.636 |
ICSI |
19.4(71) |
18(66) |
|
No. of retrieved oocytes, median (IQR) |
15(11–20) |
15(10–20) |
0.513 |
No. of thawed embryo, % (n) |
|||
1 |
23.2(85) |
23.5(86) |
0.930 |
2 |
76.8(281) |
76.5(280) |
|
Stage of transferred embryo, % (n) |
|||
D3 |
30.3(111) |
30.3(111) |
0.332 |
D5 |
48.4(177) |
52.5(192) |
|
D6 |
21.3(78) |
17.2(63) |
|
RT-37℃, 37°C and then room temperature; All-37℃, 37°C in all steps; IQR, interquartile range; BMI, body mass index; AMH, anti-Müllerian hormone; HRT, hormone replacement therapy; GnRHa-HRT, gonadotropin-releasing hormone agonist and hormone replacement therapy; Gn, gonadotropin; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm microinjection; D3, Day 3; D5, Day 5; D6, Day6 |
Table 2 summarizes the recovery rates and clinical outcomes between the 37°C-RT and all-37°C groups. The survival rates were not significantly different between the 37°C-RT and all-37°C groups [98.8% (638/646) versus 98.3% (636/647), P = 0.490]. However, the CPR and IR were significantly higher in the all-37°C group than in the 37°C-RT group [CPR: 67.8% (248/366) versus 58.5% (214/366), P = 0.009; IR: 52.4% (334/638) versus 45.8% (291/636), P = 0.019]. The eMR was not significantly different between the two groups [7.3% (18/248) versus 11.7% (25/214), P = 0.103].
Outcomes |
37℃-RT |
All-37℃ |
P-value |
---|---|---|---|
Survival rate, % (n/n) |
98.3(636/647) |
98.8(638/646) |
0.490 |
CPR, % (n/n) |
58.5(214/366) |
67.8(248/366) |
0.009 |
eMR, % (n/n) |
11.7(25/214) |
7.3(18/248) |
0.103 |
IR, % (n/n) |
45.8(291/636) |
52.4(334/638) |
0.019 |
RT-37℃, 37°C and then room temperature; All-37℃, 37°C in all steps; CPR, clinical pregnancy rate; eMR, early miscarriage rate; IR, implantation rete |
The results of the subgroup analysis are shown in Tables 3 and 4. When thawed cleavage embryos were transferred, the CPR and IR were non-significantly higher in the all-37°C group than in the 37°C-RT group [CPR: 48.6% (54/111) versus 39.6% (44/111), P = 0.176; IR: 33.2% (66/199) versus 28.6% (57/199), P = 0.329]. However, for thawed blastocysts, the CPR and IR were significantly higher in the all-37°C group than in the 37°C-RT group [CPR: 76.1% (194/255) versus 66.7% (170/255), P = 0.019; IR: 61% (268/439) versus 53.5% (234/437), P = 0.025]. The survival rates and eMRs for both frozen cleavage-stage embryos and blastocysts were not significantly different between the two groups (P > 0.05).
Outcomes |
37℃-RT |
All-37℃ |
P-value |
---|---|---|---|
Survival rate, % (n/n) |
97.5(199/204) |
97.5(199/204) |
1 |
CPR, % (n/n) |
39.6(44/111) |
48.6(54/111) |
0.176 |
eMR, % (n/n) |
13.6(6/44) |
7.4(4/54) |
0.311 |
IR, % (n/n) |
28.6(57/199) |
33.2(66/199) |
0.329 |
RT-37℃, 37°C and then room temperature; All-37℃, 37°C in all steps; CPR, clinical pregnancy rate; eMR, early miscarriage rate; IR, implantation rete |
Outcomes |
37℃-RT |
All-37℃ |
P-value |
---|---|---|---|
Survival rate, % (n/n) |
98.6(437/443) |
99.3(439/442) |
0.316 |
CPR, % (n/n) |
66.7(170/255) |
76.1(194/255) |
0.019 |
eMR, % (n/n) |
11.2(19/170) |
7.2(14/194) |
0.189 |
IR, % (n/n) |
53.5(234/437) |
61(268/439) |
0.025 |
RT-37℃, 37°C and then room temperature; All-37℃, 37°C of all steps; CPR, clinical pregnancy rate; eMR, early miscarriage rate; IR implantation rete |
As the protocol recommended by the vast majority of currently available thawing kits involves the thawing of vitrified oocytes/embryos/blastocysts at 37°C and RT, which has resulted in successful outcomes [36–38], few studies have compared the advantages and disadvantages of this protocol with those of other temperature protocols. To the best of our knowledge, this is the first study to assess the effects of two thawing procedures (all-37°C versus 37°C-RT) using the same thawing kit on the clinical outcomes of FET cycles. The findings of this large retrospective case-control matched study suggest that thawing at all-37°C significantly improved the clinical outcome of FET cycles compared with thawing at 37°C-RT.
Several studies have examined various thawing temperatures and durations in attempts to increase the survival and developmental potential of frozen-thawed embryos. When thawing mouse single-cell embryos, warming in 37°C sucrose solution dramatically increased embryo survival rate and development rates of 2-cell embryos and blastocysts compared to warming in 23°C air and 23°C 0.5 M sucrose solution [30]. Our results showed there were similar survival rates in the all-37°C and 37°C-RT groups, which may be attributable to the high-similarity recovery rates of the two thawing procedures (98.8% versus 98.3%). When oocytes were thawed in four steps with 2.5-min or 5-min intervals, no significant difference was observed between the two groups in terms of normal morphology rates. However, thawing with 2.5-min intervals led to a significant improvement in developmental potential compared with thawing with 5-min intervals [39]. In a previous study in which cryopreserved mouse embryos/blastocysts were thawed using the Kitazato Vitrification Kit, embryos were first warmed at 37°C for 1 min in TS, followed by 3 min in DS, and then 5 min in WS. This procedure significantly improved the rate of cooling and warming, and blastocyst development capability, compared with the 37°C-RT protocol. Furthermore, the rates of pregnancies and development to live offspring of embryos thawed using this procedure was similar to that of fresh embryos [40]. Consistent with these findings, the all-37°C thawing protocol in our study also yielded significantly increased CPR and IR compared with the 37°C-RT protocol.
Although combinations of cryoprotectants with less toxicity to cells can be used, a temperature and duration increase during thawing tends to increase cells’ permeability and thus their susceptibility to the toxicity osmotically damaging effects of cryoprotectants [41], which eventually affects embryo development. Additionally, the standard temperature at which embryos should be cultured in vitro is 37°C, as higher or lower temperatures have been shown to impair embryos’ developmental capacity [42, 43]. Compared with the 37°C-RT protocol, our warming protocol of all-37°C uses standard embryo in vitro culture temperatures and shortens the time of exposure in DS and WS. Thus, the damaging effects of hypothermia on embryos were minimized.
Our study has several strengths. First, we adopted strict inclusion and exclusion criteria to exclude women with potential confounding characteristics that might have influenced the FET outcomes, excluded poor-quality embryos/blastocysts and adopted a large sample size, all of which strengthen the reliability of our results. Second, all of the embryo/blastocyst quality evaluation, freezing, and thawing operations were performed according to standard protocols by two experienced embryologists, which reduced operational and observational bias. Third, a 1:1 case-control matching approach was adopted, which allowed us to control for the effects of potential confounding factors.
However, our study has some limitations. First, this was a retrospective study, which may have introduced unavoidable selection and confounding bias. Second, our study spanned a relatively long period, during which the operations of different steps may have been inconsistent and thus affected our results. Finally, our results were based on data from only a single reproductive center and a limited sample size. Therefore, larger sample sizes from multiple centers should be adopted in future studies to support our findings.
Our findings suggested that compared with the traditional 37°C-RT thawing protocol for vitrified embryos, the all-37°C thawing protocol for vitrified embryos/blastocysts improves the clinical outcomes of FET cycles and that this may be due to reduced exposure to the DS and WS in the latter protocol. A multicenter prospective randomized controlled trial is required to confirm the advantages of the all-37°C thawing protocol.
Funding This study was supported by Kunming Medical University (No. CXTD202105).
Conflict of Interest The authors declare no competing interests.
Author contributions SZ, GY, and YY designed the study. WY, LL, LW, and DZ collected the data. SZ and GY performed the data analyses and wrote the manuscript. All authors approved the final manuscript.
Ethics approval and consent to participate This study was conducted according to the Declaration of Helsinki and approved by the Ethics Committee of the First Affiliated Hospital of Kunming Medical University, and no informed consent was required because the study was a retrospective study.