Comparison of middle cerebral artery occlusion models conducted by Koizumi and Longa methods: A systematic review and meta-analysis of rodent data

doi:10.21203/rs.3.rs-2398116/v1

Ischemic stroke in rodents is usually induced by intraluminal middle cerebral artery occlusion (MCAO) via common carotid artery (CCA) plugging filament invented by Koizumi et al (MCAO-KM), or external carotid artery (CCA) plugging filament created by Longa et al (MCAO-LG). To date, a systematic comparison between the two methods remains missing. Here, we performed a meta-analysis in terms of model establishment, cerebral blood flow (CBF), and cerebral ischemia-reperfusion injury (CIRI) between of them. Literature mining suggests that MCAO-KM brings shorter operation time (p = 0.007), higher probability of plugging filament (p < 0.001) and molding establishment (p = 0.006), lower possibility of subarachnoid hemorrhage (SAH) (p = 0.02), larger infarct volume (p = 0.003), and severer brain edema (p = 0.002) and neurological deficit (p = 0.03). Nevertheless, MCAO-LG shows more adequate CBF after ischemia-reperfusion (p < 0.001), higher model survival rate (p = 0.02), and greater infarct rate (p = 0.007). In conclusion, the MCAO-KM method is simple to operate with high modeling success rate, and it is suitable for the study of brain edema under long-term hypoperfusion, the MCAO-LG method is highly challenging for novices, and it is suitable for the study of CIRI caused by acute ischemia-reperfusion. These findings are expected to benefit in the selection of intraluminal filament MCAO models prior to undertaking ischemic stroke preclinical effectiveness trials.

Middle cerebral artery occlusion

Koizumi's method

Longa's method

Animal models

Cerebral blood flow

Cerebral ischemia reperfusion injury

Stroke is one of the leading causes of death and long-term disability across the world, and a reliable ischemia stroke model is essential to produce an effective therapy for the destructive cerebrovascular accident[1–3]. The MCAO model conducted by intraluminal filament is considered as the most clinically relevant surgical model to mimic human ischemic stroke with advantage of minor craniotomy trauma, stable controllability of reperfusion, and high reproducibility, etc[4, 5]. This model was first reported by Koizumig in 1986[6], and modified by Longa in 1989[7], it have produced a profound influence on the research of ischemic stroke.

The main differences between Koizumig's method (MCAO-KM) and Longa's method (MCAO-LG) reside in the route of filament insertion to the cerebral artery and, subsequently, the CBF resupply mode as well as degree of reperfusion. Specifically, for MCAO-LG, the occlusion of middle cerebral artery (MCA) by filament is accomplished via inserting the external carotid artery (ECA), and the ischemic tissue is reperfused by bilateral common carotid arteries (CCA). But, for MCAO-KM, the filament blocked MCA is achieved by plugging through the CCA, and the reperfusion is accomplished via contralateral CCA by means of Willis ring (Fig. 1a).

The intraluminal filament model is applicable to the study of preclinical neuroprotective drugs. Some scholars have conducted comparative studies on the two methods, but the results of previous studies were rooted in a relatively small sample size, and there is no relevant meta-analysis to systematically compare their distinctions. Methodically reviewing and meta-analysis of all relevant studies in an objective and quantitative manner provide us with much credible and solid evidence to demonstrate the unique characteristics of each method. Therefore, in this study, we conducted a meta-analysis to determine the distinctions of modeling establishment and brain injury (Fig. 1b) between the two methods, aiming to provide reference for practitioners in this field.

2.1 Search strategy and study inclusion

The review was registered on the PROSPERO database prior to initiation (registration number CRD42022331652). A comprehensive search strategy was conducted in Web of Science, PubMed, Chinese National Knowledge Infrastructure (CNKI) and Wanfang database from their inceptions to July 2022 with the search terms: (Longa's method OR external carotid artery insertion OR bilateral common carotid reperfusion) AND (Koizumi's method OR common carotid artery insertion OR unilateral common carotid reperfusion) AND (intraluminal filament model OR middle cerebral artery occlusion OR focal cerebral ischemia).

We included articles if (1) animal models: MCAO model was built by intraluminal filament method, without restriction of ischemia and reperfusion duration; (2) interventions: the route of filament into the brain (common carotid artery insertion and external carotid artery insertion), or a surgical method (Koizumi's method and Longa's method), or a reperfusion mode (bilateral common carotid reperfusion or unilateral common carotid reperfusion); (3) Outcomes: operation duration, inserting filament success rate, probability of SAH, model animal mortality rate and modeling success rate, cerebral infarction size, brain edema rate, neurological deficit score.

2.2 Data Extraction and quality assessment

Two independent reviewers assessed related articles for the eligibility, and extracted the following details: (1) document elements: first author and year of publication; (2) animal datas: strains, sex, and weight (or age); (3) surgical details: anesthetic, ischemia duration, and reperfusion duration; (4) experimental outcomes: mean value, standard deviation, and sample size. WebPlotDigitizer (Version 4.4. 2020) was used for data extrapolation from graphs of published articles. In addition, any divergences were resolved by a senior author.

Quality of evidence in included studies was conducted based on a ten-item modified scale[9, 10]: (a) peer reviewed publication; (b) random allocation; (c) control of physiological parameter; (d) blinded conduct of the experiments; (e) blinded assessment of outcome; (f) use of anesthetic without significant neuroprotective activity; (g) complications in animals (aged, diabetic and hypertensive etc.); (h) sample size calculation; (i) compliance with animal welfare regulations; (j) statement of potential conflict of interests.

2.3 Statistical analysis

RevMan (version 5.4) was used for meta-analysis. Relative risk (RR) was estimated for all dichotomous variables. The weighted mean differences (WMD or MD) was calculated as a summary statistic if the continuous-type variable outcomes adopted the same scale, and the standardized mean difference (SMD) was used if the indexes were measured by various methods or techniques. Heterogeneity between studies was assessed via a standard chi-square test and I² statistic, p ≥ 0.01, I² ≤ 50% and p < 0.01, I² > 50% are considered low and high heterogeneity, respectively. If no statistical evidence of heterogeneity existed, the fixed effect (FE) model was performed with 95% confidence interval (CI); if statistical heterogeneity was found, the sensitivity analysis and subgroup analysis were used. Statistically significance for all analyses were considered if p < 0.05. Egger’s tests were employed to detect publication bias, which was completed by Stata (version 15.1).

3.1 Study search and selection

After initial search from four English databases (Web of Science and PubMed) and Chinese databases (CNKI and Wanfang), a total of 142 potentially eligible studies were identified, and 61 studies were ruled out due to duplication and irrelevance. Further, thirty-nine studies were excluded owing to invalid records, and 14 studies were removed from the remaining 42 full-text articles post careful investigation. Ultimately, a total of 28 studies were included in the systematic mete-analysis. The flowchart of literature search is provided in the Supplement 1.

3.2 Study Characteristics

All the experimental animals were male-dominated rodents, except for one study sexes in half. Isoflurane were frequently used as anesthetics, others, including ketamine, pentobarbital sodium and fentanyl, etc. The treatment and plugging depth of the filament were reported in almost all literature. Cerebral ischemic injury in the included studies was induced by transient MCAO (tMCAO) or permanent MCAO (pMCAO). The characteristics of the included studies are provided in the Supplement 2.

The comparison of model establishment, CBF and ischemic brain injury were the focus of our study. Among them, the success rate of plugging filament, incidence of SAH, model mortality rate and model success rate were analyzed as dichotomous variables. While, the surgical operation duration, cerebral infarction size, brain edema rate and neurological deficit score were analyzed as continuous variables. The calculation formula of relevant indicators are shown in Supplement 3.

3.3 Quality of Included Studies

Study quality scores for 28 studies counted in this meta-analysis were summarized in Table 1. The quality scores varied from 3 to 8 with the average as 5.46. All the studies were peer-reviewed publications. Twenty-four studies declared randomization of group allocation, and 21 studies described the monitoring of physiological parameters. Besides, one study masked the details of experimental designs, and 14 studies reported blinded assessment of outcome. Thirteen studies avoided the use of anesthetics with marked neuroprotective properties. None of the studies reported the application of co-morbidities in animals. Twenty-five studies reported sample size calculation. Among all of them, 18 studies stated the compliance with animal welfare regulations, and 9 studies addressed the conflict of interests (Table 1).

Table 1

Quality evaluation of included studies.
Year	Lead author	A	B	C	D	E	F	H	I	J	Total
1993	Laing,R.J. [11]	+		+				+			3
2000	Qu Qiumin [12]	+	+					+			3
2001	Cao Yongjun [13]	+	+	+			+	+			5
2004	Xi Gangming [14]	+	+	+		+		+			5
2006	Sun Guobing [15]	+	+				+	+			4
2006	Johannes Woitzik [16]	+	+	+		+			+	+	6
2007	Hiroaki Sakai [17]	+	+	+		+		+	+		6
2008	Cam Ertugrul [18]	+		+			+	+	+		5
2008	Sun Yu [19]	+	+				+	+			4
2008	Xiang Heng [20]	+	+				+	+	+		5
2008	Yang Zanzhang [21]	+	+	+			+	+			5
2009	Yang Debing [22]	+	+					+			3
2011	Trueman, R, C. [23]	+	+	+	+	+		+	+		7
2012	Liu Jianren [24]	+	+	+		+	+	+	+	+	8
2012	Zhao Kai [25]	+	+				+	+			4
2013	Tang Qiqiang [26]	+	+	+			+	+	+	+	7
2014	Fan Ruijuan [27]	+	+			+	+	+			5
2015	Morris,G.P. [28]	+	+	+		+		+	+	+	7
2015	Smith,H.K. [29]	+		+		+			+		4
2016	Zheng Jianfeng [30]	+	+	+			+	+	+		6
2016	Zuo Xialin [31]	+	+	+		+	+	+	+		7
2016	Cai Qiang [32]	+		+		+			+	+	5
2017	Melissa,T.L. [33]	+	+	+		+		+	+	+	7
2019	Wang Dongliang [34]	+	+	+				+	+		5
2021	Onufriev, M.V. [35]	+	+	+		+		+	+	+	7
2022	Wang Wenxiu [36]	+	+	+			+	+	+		6
2022	Yang Zhong [37]	+	+	+		+		+	+	+	7
2022	Helena, Justic [38]	+	+	+		+		+	+	+	7

Study quality items are A, Peer-reviewed publication; B, Random grouping; C, Monitoring of physiological parameters (eg. temperature, blood pressure, gases); D, Blinded conduct of ischemia; E, Blinded assessment of outcomes; F, Use of anesthetic without marked intrinsic neuroprotective properties (eg, isoflurane, ketamine, halothane ); G, Animal with co-morbidities (eg., diabetic, aged, or hypertensive); H, Sample size calculation; I, Statement of compliance with animal welfare regulations; J, Statement of potential conflict of interests.

3.4 Comparisons between the two methods in the model preparation

3.4.1 Distinction on operation duration

Four studies[19, 27, 28, 32]reported the duration of surgical operation, and one[28] was excluded due to absence of extractable date. Overall, the operation duration of MCAO-KM was significantly shorter than that of MCAO-LG (MD=-10.72, 95% CI [-18.58, -2.86], p = 0.007, heterogeneity I² = 98%, p < 0.00001) (Fig. 2a). The separation and ligation of the pterygopalatine artery (PPA) as well as the surgical proficiency may be important sources of heterogeneity. Of note, the study[28] also point out that the operation time of MCAO-KM was shorter than MCAO-LG methods, which consistent with the pooled MD estimation.

3.4.2 Distinction on probability of plugging filament

The success rates of filament insertion were reported in three studies[11, 22, 27]. Overall, the MCAO-KM shows a higher probability of plugging filament than MCAO-LG (RR = 1.43, 95% CI [1.20, 1.70], p < 0.0001, heterogeneity I² = 84%, p = 0.002) (Supplement 4), and one study [27] should be the source of heterogeneity through sensitivity analyses. After removal of this study, the pooled result between the two methods remained similar and the heterogeneity was decreased (RR = 1.59, 95% CI [1.30, 1.96], p < 0.00001, heterogeneity I² = 0%, p = 0.53) (Fig. 2b).

3.4.3 Distinction on postoperative mortality

Nineteen studies[14, 18–24, 27–32, 34–38] reported the rodent mortality after plugging filament, and one study[37] was excluded due to lack of extractable data. Overall, and, no significant difference between two methods was observed (RR = 1.09, 95% CI [0.86, 1.38], P = 0.47, heterogeneity I² = 21%, P = 0.20) (Fig. 3). Of note, sub-analyses show that model animals produced by MCAO-KM exhibited higher mortality than that by MCAO-LG in the tMCAO subgroup (RR = 1.44, 95%CI [1.07, 1.95], p = 0.02, heterogeneity I² = 0%, p = 0.73), whereas, the opposite result was found in the pMCAO subgroup (RR = 0.66, 95%CI [0.45, 0.96], p = 0.03, heterogeneity I² = 32%, p = 0.20). The p-value of Egger's regression test in the tMCAO subgroup was 0.291, indicating no significant publication bias (Supplement 5).

3.4.4 Distinction on occurence of SAH

Nine studies [18, 19, 21, 24–27, 31, 38] reported the occurrence of SAH after operation, and one study [26] was excluded due to lack of detailed data. Overall, the meta-analysis results indicate that the modeling method of MCAO-KM causes a lower probability of SAH compared to the MCAO-LG with no substantial heterogeneity (RR = 0.43, 95% CI [0.20, 0.90], p = 0.02, heterogeneity I² = 0%, p = 0.96) (Fig. 4).

3.4.5 Distinction on success rate of MCAO models

The success rates of MCAO models were assessed in 16 studies [12, 18–28, 30, 34–36]. Overall, the MCAO-KM brings statistically higher probability on model success than the MCAO-LG (RR = 1.14, 95% CI [1.03, 1.27], p = 0.01, heterogeneity I² = 45%, p = 0.02). Subgroup analyses suggest that the MCAO-KM supports higher model success rate in the pMCAO subgroup (RR = 1.34, 95% CI [1.09, 1.66], p = 0.006, heterogeneity I² = 0%, p = 0.51), but not tMCAO (RR = 1.06, 95% CI [0.94, 1.20], p = 0.33, heterogeneity I² = 46%, p = 0.04) (Supplement 6a).

Sensitivity analysis of the tMCAO subgroups revealed one study [25] may be the source of heterogeneity. After deletion of this study, no significant difference between the two methods was found in overall effect (RR = 1.08, 95% CI [0.97, 1.20], p = 0.14, heterogeneity I² = 24%, p = 0.17) as well as tMCAO subgroup (RR = 0.97, 95% CI [0.86, 1.10], p = 0.64, heterogeneity I² = 0%, p = 0.53). However, the MCAO-KM method's success rate remains higher in the pMCAO subgroup (Fig. 5). The p-value of Egger's regression test in the tMCAO subgroup was 0.089, indicating no significant publication bias (Supplement 6b & 6c).

3.5 Comparisons between the two methods in CBF and brain injury

3.5.1 Distinction on CBF after ischemia or ischemia-reperfusion

Eight studies [13, 16, 28, 29, 32, 33, 37, 38] with 9 comparisons were applied to evaluate CBF after cerebral ischemia or ischemia-reperfusion between MCAO-KM and MCAO-LG, and one study [32] was excluded due to lack of sample size, and another study [16] was also excluded because it did not reperfused by removing the filament. Overall, there was no significant difference in CBF between the two methods during ischemia (SMD = 0.18; 95% CI [-0.32, 0.68], p = 0.48; heterogeneity I² = 40%, p = 0.14) (Fig. 6a). Interestingly, after reperfusion, the CBF achieved by MCAO-KM was dramatically lower than that by MCAO-LG (SMD=-1.34; 95% CI [-1.85, -0.83], p < 0.00001; heterogeneity I² = 60%, p = 0.02) (Supplement 7). After exclusion of one study [38] as a potential source of heterogeneity, the CBF outcomes between the two methods remained similar and the heterogeneity was significantly reduced (SMD=-1.12; 95% CI[-1.65, -0.59], p < 0.0001; heterogeneity I² = 11%, p = 0.35) (Fig. 6b).

3.5.2 Distinction on cerebral infarction size

Eleven studies [13, 15, 19, 21, 27–29, 31, 34, 35, 39] evaluated the cerebral infarction rates between MCAO-KM and MCAO-LG methods (Supplement 8༆Supplement 9), and one study [34] was excluded due to lack of sample size. Overall, the cerebral infarction rates of the MCAO-KM was markedly lower than the MCAO-LG (MD=-3.37; 95% CI[-4.71, -2.04], p < 0.00001; heterogeneity I² = 82%, p < 0.00001) (Supplement 8a). Sub-analyses suggest that the MCAO-KM gets a low cerebral infarction rates in tMCAO subgroup, but not in pMCAO subgroup (MD=-3.53, 95% CI [-4.91, -2.15], p < 0.00001, heterogeneity I² = 84%, p < 0.00001; and MD=-1.12; 95% CI [-6.37, 4.14]; p = 0.68 heterogeneity I² = 0%, p = 0.95, respectively) (Supplement 8a). After removal of one study[29], the distinction of cerebral infarction rates remained significant between the two methods, and the heterogeneity of overall effect as well as tMCAO subgroup was eliminated (SMD=-2.10, 95% CI [-3.63, -0.57], p = 0.007, heterogeneity I² = 39%, p = 0.10 in tMCAO subgroup; and SMD=-2.02, 95% CI [-3.49, -0.56], p = 0.007, heterogeneity I² = 26%, p = 0.18 in overall effect) (Fig. 7a). The p-value of Egger's regression test in the tMCAO subgroup was 0.390, verifying no significant publication bias (Supplement 8b ༆ 8c).

Twelve studies [13, 14, 16–18, 20, 24, 30, 32, 33, 37, 38] evaluated the distinction of cerebral lesion volumes between the two modeling methods (Supplement 10༆Supplement 11). Meta-analysis suggests that the average cerebral lesion volumes of the MCAO-KM rodents was higher than the MCAO-LG rodents (SMD = 0.68; 95% CI [0.46, 0.91], p < 0.00001; heterogeneity I² = 59%, p = 0.003) (Supplement 10a). Sub-analyses suggest that the MCAO-KM animals show a higher average cerebral infarction volume than MCAO-LG in both tMCAO and pMCAO subgroups (SMD = 0.61; 95% CI [0.36, 0.86], p < 0.00001; heterogeneity I² = 57%, p = 0.003; and SMD = 0.96, 95% CI [0.47, 1.45], p = 0.0001, heterogeneity I² = 66%, p = 0.001, respectively) (Supplement 10a). Sensitivity analyses pointed out that two studies [30, 38] should be the source of heterogeneity. After removing these studies, the heterogeneity was significantly reduced. Sub-analyses suggest that the MCAO-KM animals showed a higher average cerebral lesion volume in tMCAO but not pMCAO subgroup (SMD = 0.54; 95% CI [0.28, 0.80]; p < 0.0001, heterogeneity I² = 0%, p = 0.84 in tMCAO; SMD = 0.57, 95%CI [0.00, 1.13], p = 0.05, heterogeneity I² = 39%, p = 0.16 in pMCAO; and SMD = 0.54, 95%CI [0.31, 0.78], p < 0.00001, heterogeneity I² = 0%, p = 0.67 in overall effect) (Fig. 7b). The p-value of Egger's regression test in the tMCAO subgroup was 0.679, indicating no significant publication bias (Supplement 10b & 10c).

3.5.3 Distinction on brain edema

The assessments of the brain edema rates were performed in six studies[12, 16, 24, 26, 34, 38], and two studies [26, 34] were excluded due to lack of detailed data (Supplement 12). Overall, the MCAO-KM animals produced severer brain edema compared to the MCAO-LG with marked heterogeneity (SMD = 0.53, 95% CI [0.03, 1.02], p = 0.04, heterogeneity I² = 71%, p = 0.005) (Supplement 12). Subgroup analysis indicates that significant differences occurs in the tMCAO but not in the pMCAO subgroup (SMD = 0.70, 95% CI [0.12, 1.28], p = 0.02, heterogeneity I² = 78%, p = 0.003, and SMD = 0.10, 95% CI [-0.82, 1.02], p = 0.83, heterogeneity I² = 51%, p = 0.15, respectively). Sensitivity analysis uncovered that one study [38] is a potential source of heterogeneity. After removal of this study, the pooled estimation of brain edema between the two MCAO animals remained similar and the heterogeneity was eliminated (SMD = 0.93, 95% CI [0.34, 1.53], p = 0.002, heterogeneity I² = 0%, p = 0.88 in tMCAO subgroup; SMD = 0.10, 95% CI [-0.82, 1.02], p = 0.83, heterogeneity I² = 51%, p = 0.15 in pMCAO subgroup; and SMD = 0.68, 95% CI [0.18, 1.18], p = 0.007, heterogeneity I² = 12%, p = 0.34 in overall effect) (Fig. 8).

3.5.4 Distinctions on neurological deficit scores

Nineteen studies evaluated neurological deficits via Longa score[13, 16, 19, 22, 25, 26, 30–32, 34, 36], Bederson score[24, 35], modified neurological severity scores (mNSS) score[17, 29, 37], and Garcia score[24, 27, 38] (Supplement 13). Among them, three articles [16, 26, 32] were excluded due to absence of available data. Of note, the higher the Garcia score, the better the neurobehavior, which is contrary to other scoring methods, thus, these studies using Garcia score were not pooled into the sub-analysis (Supplement 14). Overall, the average neurological deficit scores of the MCAO-KM animals significantly higher than that of MCAO-LG animals without heterogeneity (SMD = 0.21, 95% CI [0.02, 0.39], p = 0.03, heterogeneity I² = 20%, p = 0.21) (Fig. 9).

Stroke accounts for 11.1% of global deaths and remains a worldwide health burden [40]. About 87% of strokes in humans are ischemic strokes, and 70% of cerebral infarcts are caused by occlusion of the MCA and its branches[41]. The MCAO-KM and MCAO-LG modeling methods are the two most common ways of achieving MCAO as classical surgical approaches in preclinical experiments, and they are invaluable for researchers to understand stroke patho-physiology and develop new therapeutics. However, in preclinical studies, the two classic intraluminal filament approaches for MCAO are considered alternatives, and some researchers even believed that they are interchangeable [24, 31]. Therefore, it is critical to distinguish between them in order to develop an appropriate stroke model.

4.1 Distinctions in model establishment

The fundamental distinction between the MCAO-KM and MCAO-LG methods is the surgical procedure's complexity. Because most anesthetics have neuroprotective benefits against ischemic stroke, the prolonged duration of anesthesia may impact the therapeutic effect of candidate medications [42]. According to our experience, the MCAO-KM is easier to perform because it does not require ligating the ECA and its embranchment prior to ischemia, which shortens operation time (Fig. 2a) and may decrease eating difficulties caused by injuries of soft tissues and cranial nerves [43]. Furthermore, CCA is easier to plug a filament than ECA because the former is thicker and straighter than the latter (Fig. 2b), which explains the reason why MCAO-KM shows higher success rate of insertion filament and shorter operation time than MCAO-LG.

SAH is one of the leading causes of postoperative mortality in intraluminal filament MCAO model, usually caused by excessive insertion of the filament [44]. Generally, the optimal insertion depth depends on the species, and the propulsion of the thread stops when the finger feels a slight resistance after it enters the intracranial cavity [29, 32]. Arterioles derive from the ECA are prone to bleeding, thus, the slipknots around ECA need to be tied more tightly (Fig. 1a), which results in a reduced perception of protraction resistance by the fingers. It may be a direct factor causing elevated SAH in MCAO-LG animals with a higher risk of intracranial vascular puncture (Fig. 4).

Low mortality is advantageous, representing an improved model of MCAO, allowing reduced number of rodents to be used to meet the research goal. There was no overall difference in model mortality between the two surgical methods, but the distinct results were found in the subgroup analyses (Fig. 3). In the tMCAO subgroup, the animal mortality of the MCAO-KM was higher than MCAO-LG (Fig. 3a), which may be related to the long-term hypoperfusion of the ischemic hemisphere caused by permanent ligation of CCA [29]. Conversely, in the pMCAO subgroup, the higher mortality occurs in MCAO-LG (Fig. 3b), which may be interpreted by the greater surgical damage caused by a relatively complex procedure.

Model success rate, as a comprehensive index, reflects the difficulty of operation. For the tMCAO model, there is no significant difference between the two methods (Fig. 5a), however, for the pMCAO model, the success rate of MCAO-KM is higher than MCAO-LG (Fig. 5b).Given the foregoing, it is reasonable to believe that the MCAO-KM method is simple to model establishment. In other words, the MCAO-LG method is challenging for a fledgling.

4.2 Distinctions in CBF

The mode and extent of reperfusion are the main distinctions between the two methods in physio-pathology. In short, the MCAO-LG animals' reperfusion is mainly derived from bilateral CCA, whereas the MCAO-KM animals' reperfusion is primarily derived from contralateral CCA via Willis ring because the ipsilateral CCA was permanently ligated (Fig. 1a). After plugging in the filament, CBF detection revealed that no significant difference between the two groups (Fig. 6a), and the decline was all about 60% of the pre-ischemia level [16, 18]. Interestingly, reperfusion in MCAO-KM animals reached only 50% of baseline levels, while, in MCAO-LG rodents, reperfusion could rapidly restore to near-normal levels (Fig. 6b). Furthermore, high-resolution MR angiography defines the MCAO-KM as a ischemia-reperfusion model with chronic hypoperfusion, whereas the MCAO-LG method achieves ischemia complete reperfusion [38].

Surge reperfusion observed with removal of the filament may be like that observed with clinical endovascular thrombectomy[45]. Considering the insufficient reperfusion of MCAO-KM animals, some scholars believe that the MCAO-LG method with complete reperfusion should be given priority in the exploration of the neuroprotective agents on cerebral ischemia-reperfusion injury [45]. In fact, only 7–10% patients in developed countries receive effective thrombolytic therapy within the treatment time window [46, 47]. What's worse, stroke patients - approximately 30.9–72.3% - do not achieve full revascularization despite receiving thrombolytic therapy [48–50]. Besides, the researchers observed that the brain would provide hypoperfusion blood to the infarcted core via collateral vessels in humans or rodents that did not receive recanalization treatment [51, 52]. Thus, stroke patients with chronic hypoperfusion injury was more common than that of complete reperfusion [53]. In other words, although the ischemia stroke model prepared by the MCAO-KM method does not achieve complete reperfusion, it may be closes to the main situation of no or incomplete thrombolysis.

4.3 Distinctions in cerebral infarction size, brain edema rate and neurological deficit score

We evaluated the infarction size in forms of volume and percentage, and interestingly, the results from the two versions were polar opposites (Fig. 7). As is well known, full ischemia-reperfusion rapidly mobilizes amounts of erythrocytes and hemoglobin [54], and mitochondrial dysfunction causes the oxygen transported by this hemoglobin to be converted into amounts of reactive oxygen species [55, 56]. Additionally, via interacting with endothelial cells, a considerable number of circulating immune cells enter the brain parenchyma after ischemia-reperfusion [57]. All of the aforementioned factors lead to untimely oxidative stress and inflammatory reactions, which result in secondary CIRI. Another investigation discovered that rats that received quick flow restoration had more infarctions than those who underwent incremental flow restoration [58]. Thus, it is speculated that the bigger cerebral infarction in the MGO-LG rodents is caused by the more severe CIRI produced by the remarkable recovery of CBF.

Ligation of a unilateral CCA significantly destroys the blood-brain barrier and increases brain water content in rats [59], Similar to this, individuals who have failed recanalization experience an increase and expansion of the edema volume [60]. Therefore, it may be reasonable to assume that the permanent closure of CCA following filament removal explains why the MCAO-KM technique results in more severe brain edema (Fig. 8). Giving that edema-corrected infarction progression is greater than the edema progression, the level of edema is insufficient to account for the overall expansion of infarction volume [60]. When estimating how an intervention would affect infarct size, a modified formula rather than pure infarction volume must be used since the peak of cerebral edema, which happens after 2 to 3 days, should be taken into account as for calculating the infarction volume [61].

Neurological abnormalities and cognitive deterioration are linked to chronic cerebral hypoperfusion. In this study, we discovered that neurological impairments in MCAO-KM rodents were more severe as a result of the protracted hypoperfusion brought on by unilateral CCA ligation (Fig. 9). Important to keep in mind is that the impact of permanent CCA blockage on cognitive function may negate the effectiveness of emerging treatment options that aim to treat stroke-induced vascular dementia or stroke with Alzheimer's disease[37].

4.4 Distinctions in inflammation

The development of ischemia disease is heavily influenced by inflammation, and the inflammatory response brought on by various surgical techniques is time-dependent. As early as 4–6 hours after surgery, neutrophils were observed in the ischemic core following ischemia [62, 63]. The interaction between leukocytes and endothelial cells was studied by Smith H K et al [29], who discovered that both animals' ischemic brain tissue was filled with a significant number of rolling leukocytes. Interestingly, cells remaining stationary within the vessel only occurred in MCAO-LG after urgent reperfusion. Another study [34] discovered that NF-κB expression in the ischemic core was considerably greater in the MCAO-LG than the MCAO-KM after 24 hours of reperfusion. However, during the subacute stage of ischemic stroke, Onufriev et colleagues [35] discovered that IL-1 together with corticosterone discharge and amass in the MCAO-KM rat hippocampus, indicating that MCAO-KM predisposes the animals to corticosterone-dependent distant neuroinflammatory hippocampal injury. Furthermore, in terms of long-term inflammatory responses, Yang et al [37] found no significant difference between the two modeling approaches in astrocyte and microglial activation, as well as apoptotic neuronal death. Thus, MCAO-LG may have a greater inflammatory response in the acute phase, whereas MCAO-KM may have a stronger inflammatory response in the subacute period.

4.5 Distinctions in other aspects

Mice's pterygopalatine artery, the first internal carotid artery branch, is the source of the ocular artery [38]. The majority of retinal reactions to ischemia coincide with brain tissue in MCAO-KM animals, with substantial atrophy of the inner retinal layers due to permanent CCA ligation, whereas MCAO-LG mice revealed no serious histological retinal abnormalities [38]. Thus, the MCAO-KM was created to better study the pathophysiology of cerebral combine retinal ischemia.

Based on the meta-analysis results, we conclude that the surgical procedure - MCAO-KM or MCAO-LG - should not be chosen arbitrarily, and the filament insertion route has a substantial effect on model establishment, CBF, and CIRI.

In summary, MCAO-KM has advantages in model establishment, such as shorter operation time, easier filament insertion, lower risk of SAH, and higher model success rate; it is more appropriate for the study of ischemia-induced brain edema due to chronic hypo-reperfusion with a higher risk of death and severe neurological deficits. However, MCAO-LG is better suited for the study of acute CIRI due to its ability to preserve CCA integrity with a sophisticated and challenging surgical procedure, it has a low model mortality, high CBF replenishment, large cerebral infarction, and an acute inflammatory response.

Instead of arbitrary variables, this discovery may provide scientists with actual parameters for selecting a suitable intraluminal filament MCAO model.

BBB, blood-brain barrier; CBF, cerebral blood flow; CCA, common carotid artery; CIRI: cerebral ischemia -reperfusion injury; ECA, external carotid artery; ICA, internal carotid artery; MCA, middle cerebral artery; MCAO, middle cerebral artery occlusion; mNSS, modified neurological severity scores; pMCAO, permanent middle cerebral artery occlusion; PPA, pterodylopalatal artery; SAH, subarachnoid hemorrhage; tMCAO, transient middle cerebral artery occlusion.

Acknowledgements

We appreciate all researchers, some because their literature was included in our meta-analysis, and others because they were involved in this work.

Ethical Approval

This research does not require an ethics approval since no experiments with human participants or animal samples were contained as part of the review.

Competing interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Authors' contributions

Yong Li, Bowen Deng, Xianzhi Huang and Lin Liu of our team are responsible for collecting relevant documents, Yong Li and Liying He are responsible for writing the first draft of this paper, Sijing Liu, Li Tan, and Caixia Yang are responsible for reviewing the manuscript, and Yong Li is responsible for submitting the final manuscript.

Funding

This work was financially supported by grants from National Natural Science Foundation of China (81872959, 81373920).

Availability of data and materials

All data generated or analyzed during this study are included in this article and its supplementary materials.

Supplementary Information

The online version contains supplementary material available at https://

Benjamin EJ, Virani SS, Callaway CW, Chang AR, Muntner P. Heart Disease and Stroke Statistics—2018 Update: A Report From the American Heart Association. Circulation. 2018;137(12):558.
Owolabi MO, Akarolo-Anthony S, Akinyemi R, Arnett D, Gebregziabher M, Jenkins C, et al. The burden of stroke in Africa: a glance at the present and a glimpse into the future. Cardiovasc J Afr. 2015;26(2):S27.
Wu SM, Wu B, Liu M, Chen ZM, Wang WZ, Anderson CS, et al. Stroke in China: advances and challenges in epidemiology, prevention, and management. The Lancet Neurology. 2019;18(4):394-405.
Sommer CJ. Ischemic stroke: experimental models and reality. Acta Neuropathol. 2017;133(2):245-261.
Howells DW, Porritt MJ, Rewell SS, O'Collins V, Sena ES, van der Worp HB, et al. Different strokes for different folks: the rich diversity of animal models of focal cerebral ischemia. J Cereb Blood Flow Metab. 2010;30(8):1412-1431.
Koizumi J, Yoshida Y, Nakazawa T, Ooneda G. Experimental studies of ischemic brain edema. Nosotchu. 1986;8(1):1-8.
Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke. 1989;20(1):84-91.
Uluc K, Miranpuri A, Kujoth GC, Akture E, Baskaya MK. Focal cerebral ischemia model by endovascular suture occlusion of the middle cerebral artery in the rat. J Vis Exp. 2011(48).
Landis SC, Amara SG, Asadullah K, Austin CP, Blumenstein R, Bradley EW, et al. A call for transparent reporting to optimize the predictive value of preclinical research. Nature. 2012;490(7419):187-191.
Macleod MR, O Collins T, Howells DW, Donnan GA. Pooling of Animal Experimental Data Reveals Influence of Study Design and Publication Bias. Stroke. 2004;35(5):1203-1208.
Laing RJ, Jakubowski J, Laing RW. Middle cerebral artery occlusion without craniectomy in rats. Which method works best? Stroke. 1993;24(2):294-297.
Qu QM, Cao ZL, Yang JB. Comparison of Longa method and Koizumi method in focal cerebral ischemia model of middle cerebral artery occlusion in rats. Chinese Journal of Neurology. 2000;22(05):32.
Cao YJ, Cheng YB. The improvement and discussion of the model of focal cerebral ischemia/reperfusion with suture-occluded method in rats. Chinese Journal of Applied Physiology. 2001(02):95-97.
Xi GM, Wang HQ, He GH, Huang CF, Wei GY. Evaluation of murine models of permanent focal cerebral ischemia. Chin Med J (Engl). 2004;117(3).
Sun GB, Xu K, Ke X, Zhao B, Li ZH. A model of focal cerebral ischemia in rats by reversible middle cerebral artery occlusion. Stroke and Nervous Diseases. 2006(06):342-344.
Woitzik J, Schneider UC, Thomé C, Schroeck H, Lothar S. Comparison of different intravascular thread occlusion models for experimental stroke in rats. J Neurosci Methods. 2006;151(2):224-231.
Sakai H, Sheng H, Yates RB, Ishida K, Pearlstein RD, Warner DS. Isoflurane Provides Long-term Protection against Focal Cerebral Ischemia in the Rat. Anesthesiology. 2007;106(1):92-99.
Cam E, Kilic E, Yulug B, Ritz MF. Occlusion of the Middle Cerebral Artery in Rats; 2008. 52-65 p.
Sun Y, Yan GF, Jiang H, Zhu YS. A Modified Middle Cerebral Artery Occlusion Method for Establishment of Brain Reperfusion Model in Rats. Chinese Journal of Comparative Medicine. 2008(08):8-10.
Heng X. Modification of focal cerebral ischemia model in rats for long term living [硕士]: Tianjin University of Sport; 2008. 2-29 p.
Yang ZZ, Chen H. Improvement and Experience in Suture Methods for Rat Model of Focal Cerebral Ischemia. Lishizhen Medicine and Materia Medica Research. 2008(09):2189-2190.
Yang DB, Zhang LH, Ma C. Development of the middle cerebral artery occlusion model and behavioral assessment improvement. Chinese Journal of Ethnomedicine and Ethnopharmacy. 2009;18(16):27-28.
Trueman RC, Harrison DJ, Dwyer DM, Dunnett SB, Hoehn M, Farr TD. A Critical Re-Examination of the Intraluminal Filament MCAO Model: Impact of External Carotid Artery Transection. Transl Stroke Res. 2011;2(4):651-661.
Liu JR, Jensen-Kondering UR, Zhou JJ, Sun F, Feng XY, Shen XL, et al. Transient filament occlusion of the middle cerebral artery in rats: does the reperfusion method matter 24 hours after perfusion? Bmc Neurosci. 2012;13(1):154.
Zhao K, Zhang G, Jia Y, Li N. Improvement and Exploration on Local Cerebral Ischemia/Reperfusion Model in Rats by Intraluminal Suture Technique. Chinese Journal of Stroke. 2012;7(12):916-921.
Tang QQ, Han RD, Xiao H, Shi LL, Shen JL, LUN Q, et al. Role of suture diameter and vessel insertion position in the establishment of the middle cerebral artery occlusion rat model. Exp Ther Med. 2013;5(6):1603-1608.
Fan RJ, Luo YF, Sun YF. Establishment of new focal cerebral ischemia-reperfusion model in rats by suture method. Chinese Journal of Ethnomedicine and Ethnopharmacy. 2014;23(02):26-27.
Morris GP, Wright AL, Tan RP, Gladbach A, Ittner LM, Vissel B. A Comparative Study of Variables Influencing Ischemic Injury in the Longa and Koizumi Methods of Intraluminal Filament Middle Cerebral Artery Occlusion in Mice. Plos One. 2016;11(2):e148503.
Smith HK, Russell JM, Granger DN, Gavins FN. Critical differences between two classical surgical approaches for middle cerebral artery occlusion-induced stroke in mice. J Neurosci Methods. 2015;249:99-105.
Zheng JF. Study on the model of middle cerebral artery occlusion on Ischemic stroke [硕士]: Fujian Medical University; 2016. 53 p.
Zuo XL, Deng HL, Wu P, Xu E. Do different reperfusion methods affect the outcomes of stroke induced by MCAO in adult rats? Int J Neurosci. 2015;126(9):850-855.
Cai Q, Xu G, Liu JH, Wang L, Deng G, Liu J, et al. A modification of intraluminal middle cerebral artery occlusion/reperfusion model for ischemic stroke with laser Doppler flowmetry guidance in mice. Neuropsychiatr Dis Treat. 2016;Volume 12:2851-2858.
Trotman-Lucas M, Kelly ME, Janus J, Fern R, Gibson CL. An alternative surgical approach reduces variability following filament induction of experimental stroke in mice. Dis Model Mech. 2017;10(7):931-938.
Wang DL, Wang D, He TP, Jing Z, Gao X, Yuan Y. Comparison between two types of middle cerebral artery occlusion model in mice by monofilament method. Chinese Journal of Neurotraumatic Surgery(Electronic Edition). 2019;5(06).
Onufriev MV, Moiseeva YV, Zhanina MY, Lazareva NA, Gulyaeva NV. A Comparative Study of Koizumi and Longa Methods of Intraluminal Filament Middle Cerebral Artery Occlusion in Rats: Early Corticosterone and Inflammatory Response in the Hippocampus and Frontal Cortex. International Journal of Molecular Sciences. 2021;22(24):13544.
Wang WX, Pu Z, Liu Q, Ding FF, Kang XW, Gong T, et al. An Evaluation of 20-day Survival Rates of MCAO Rat Model Prepared by Modified and Traditional Suture-occluded Methods. Western Journal of Traditional Chinese Medicine. 2019;33(04):42-44.
Yang Z, Li X, Cao ZP, Wang P, Warner DS, Sheng HX. Post-ischemia common carotid artery occlusion worsens memory loss, but not sensorimotor deficits, in long-term survived stroke mice. Brain Res Bull. 2022;183:153-161.
Justić H, Barić A, Šimunić I, Radmilović M, Ister R, Škokić S, et al. Redefining the Koizumi model of mouse cerebral ischemia: A comparative longitudinal study of cerebral and retinal ischemia in the Koizumi and Longa middle cerebral artery occlusion models. Journal of Cerebral Blood Flow & Metabolism. 2022;42(11):2080-2094.
Jianyuan L, Yazhuo Z, Weicheng Y, Songbai G, Meizhen S. A comparison of two methods in making ischemia-reperfusion model of middle cerebral artery in rats. Acta Aacademiae Medicinae Qingdao Universitatis. 2008(05):382-384.
Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet. 2012;380(9859):2095-2128.
Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1,000 consecutive patients with first stroke. Stroke. 1988;19(9):1083-1092.
Archer DP, Walker AM, McCann SK, Moser JJ, Appireddy RM. Anesthetic Neuroprotection in Experimental Stroke in Rodents. Anesthesiology. 2017;126(4):653-665.
Dittmar M, Spruss T, Schuierer G, Horn M. External carotid artery territory ischemia impairs outcome in the endovascular filament model of middle cerebral artery occlusion in rats. Stroke. 2003;34(9):2252-2257.
Schmid-Elsaesser R, Zausinger S, Hungerhuber E, Baethmann A, Reulen HJ. A critical reevaluation of the intraluminal thread model of focal cerebral ischemia: evidence of inadvertent premature reperfusion and subarachnoid hemorrhage in rats by laser-Doppler flowmetry. Stroke. 1998;29(10):2162-2170.
Sutherland BA, Neuhaus AA, Couch Y, Balami JS, Buchan AM. The transient intraluminal filament middle cerebral artery occlusion model as a model of endovascular thrombectomy in stroke. Journal of Cerebral Blood Flow & Metabolism. 2016;36(2):363-369.
McMeekin P, White P, James MA, Price CI, Flynn D, Ford GA. Estimating the number of UK stroke patients eligible for endovascular thrombectomy. Eur Stroke J. 2017;2(4):319-326.
Chia NH, Leyden JM, Newbury J, Jannes J, Kleinig TJ. Determining the Number of Ischemic Strokes Potentially Eligible for Endovascular Thrombectomy: A Population-Based Study. Stroke. 2016;47(5):1377-1380.
Zhang PL, Wang YX, Chen Y, Zhang CH, Li CH, Dong Z, et al. Use of Intravenous Thrombolytic Therapy in Acute Ischemic Stroke Patients: Evaluation of Clinical Outcomes. Cell Biochem Biophys. 2015;72(1):11-17.
Dalkara T, Arsava EM. Can restoring incomplete microcirculatory reperfusion improve stroke outcome after thrombolysis? J Cereb Blood Flow Metab. 2012;32(12):2091-2099.
Lin C. Serial multimodal imaging change and its effects in acute ischemic stroke patients who achieved recanalization: Zhejiang University; 2020. 138 p.
Wang J, Lin X, Mu Z, Shen F, Zhang L, Xie Q, et al. Rapamycin Increases Collateral Circulation in Rodent Brain after Focal Ischemia as detected by Multiple Modality Dynamic Imaging. Theranostics. 2019;9(17):4923-4934.
Shuaib A, Butcher K, Mohammad AA, Saqqur M, Liebeskind DS. Collateral blood vessels in acute ischaemic stroke: a potential therapeutic target. Lancet Neurol. 2011;10(10):909-921.
Ma R, Xie Q, Li Y, Chen Z, Ren M, Chen H, et al. Animal models of cerebral ischemia: A review. Biomed Pharmacother. 2020;131:110686.
Martha SR, Collier LA, Davis SM, Seifert HA, Leonardo CC, Ajmo CT, et al. Translational Evaluation of Acid/Base and Electrolyte Alterations in Rodent Model of Focal Ischemia. Journal of Stroke and Cerebrovascular Diseases. 2018;27(10):2746-2754.
Peters O, Back T, Lindauer U, Busch C, Megow D, Dreier J, et al. Increased Formation of Reactive Oxygen Species after Permanent and Reversible Middle Cerebral Artery Occlusion in the Rat. Journal of Cerebral Blood Flow & Metabolism. 1998;18(2):196-205.
Pundik S, Xu K, Sundararajan S. Reperfusion brain injury: Focus on cellular bioenergetics. Neurology. 2012;79(Issue 13, Supplement 1):S44-S51.
Kawabori M, Yenari MA. Inflammatory responses in brain ischemia. Curr Med Chem. 2015;22(10):1258-1277.
Xu WW, Zhang YY, Su J, Liu AF, Wang K, Li C, et al. Ischemia Reperfusion Injury after Gradual versus Rapid Flow Restoration for Middle Cerebral Artery Occlusion Rats. Sci Rep. 2018;8(1):1638.
Yang W, Zhang X, Wang N, Tan J, Fang X, Wang Q, et al. Effects of Acute Systemic Hypoxia and Hypercapnia on Brain Damage in a Rat Model of Hypoxia-Ischemia. Plos One. 2016;11(12):e167359.
Konduri P, van Kranendonk K, Boers A, Treurniet K, Berkhemer O, Yoo AJ, et al. The Role of Edema in Subacute Lesion Progression After Treatment of Acute Ischemic Stroke. Front Neurol. 2021;12:705221.
Swanson RA, Morton MT, Tsao-Wu G, Savalos RA, Davidson C, Sharp FR. A Semiautomated Method for Measuring Brain Infarct Volume. Journal of Cerebral Blood Flow & Metabolism. 1990;10(2):290-293.
Wang Q, Tang XN, Yenari MA. The inflammatory response in stroke. J Neuroimmunol. 2007;184(1-2):53-68.
Perera MN, Ma HK, Arakawa S, Howells DW, Markus R, Rowe CC, et al. Inflammation following stroke. J Clin Neurosci. 2006;13(1):1-8.

No competing interests reported.

SupplementalMaterial.docx

Comparison of middle cerebral artery occlusion models conducted by Koizumi and Longa methods: A systematic review and meta-analysis of rodent data

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods