Intraprocedural continuous saline infusion lines significantly reduce the incidence of acute kidney injury during endovascular procedures for stroke and myocardial infarction: evidence from a systematic review and meta-regression

Introduction

Acute kidney injury (AKI) has an incidence of 21.6% (95% confidence interval [95% CI], 19.3 to 24.1) in the acute setting(1) It has been widely demonstrated that iodine contrast administration may increase the risk of developing AKI due to transient endothelium-dependent vasodilation, and vasoconstriction potentially critically reduces renal blood flow and induces renal ischemia (2,3). Although there are no clear differences between the routes of administration of contrast media in inducing AKI (e.g., intravenous or intra-arterial), a lower threshold value of the glomerular filtration rate was chosen for intra-arterial administration(4). However, some studies have emphasized, for the risk of developing AKI, patient comorbidities, including diabetes and preexisting chronic kidney disease(2,4). However, all these studies presented several confounding factors for the case mix included(4).

Patients with acute conditions were more prone to develop AKI (incidence 21.6% [95% confidence interval [95% CI], 19.3 to 24.1]). Ischemic stroke (IS) and acute myocardial infarction (AMI) have several clinical similarities, including pathophysiology, acute onset, prognosis, and effective treatment (5). Furthermore, both diseases can be treated endovascularly using mechanical thrombectomy (MT) for IS or percutaneous coronary intervention (PCI) for AMI(6,7). Considering that both clinical presentations have the same setting, including acute disease and intra-arterial contrast administration, these patients are prone to develop AKI(6,7).

To avoid distal embolization, neurointerventions are performed under continuous saline infusion (CSI) (8), which is not mandatory for PCI(9). During MT, a continuous saline infusion line (CSIL) is connected to all catheters used during the procedure. Although it is difficult to evaluate the amount of saline administered during MT, it can be estimated that approximately 361.8 to 1085.4 ml depending on the number of catheters used (e.g., 2 co-axial catheter 2 saline infusion) and time of the procedure(10,11). Interestingly, fluid expansion is one of the main strategies used to prevent and treat AKI (12). Furthermore, the amount of saline infused during MT is similar to that recommended by the European Society of Urogenital Radiology (ESUR) to prevent and treat AKI(4,13).

This systematic review, meta-analysis, and meta-regression aimed to evaluate the impact of CSIL on the incidence of AKI during endovascular procedures for the treatment of acute diseases. Therefore, we compared the incidence of AKI after MT with that after PCI for AMI.

Material and methods

The study protocol is available upon reasonable request from the corresponding author. In addition, the data used for the systematic review were publicly available. The study was drafted according to the PRISMA guidelines(14). The review protocol was registered in PROSPERO (ID 433497). The search strategy for the MT subgroup was discussed in the following paragraph. The results were compared with those of Lun et al.(15) on the incidence of AKI in PCI on 1.2 millions of patients in 2021.

Results

Search results

Our systematic search yielded 3317 articles. A total of 57/3317 (1.7%) were automatically deleted by the Rayyan AI for duplications. A total of 97/3260 (3.0%) studies were manually excluded because of duplications. 3136/3163 (99.1%) were excluded after reviewing the title and abstract. After retrieving the full text, 9/27 (33.3%) papers were excluded. The final population for stroke treatment included 18 studies (Figure 1).

• Download figure
• Open in new tab

Figure 1. Flowchart of the mechanical thrombectomy studies subgroup

Concerning PCI, the systematic review by Lun et al. included 133 studies(15). However, this study considered both acute and elective cases. To reduce clinical imbalance, only acute PCI setting were retrieved and included (33/133 studies [24.8%]).

The study characteristics and risk of bias assessments for both MT and PCI are summarized in Tables 1 and S1.

Clinical results

A total of 69,464 patients were included in the analysis, including 30,138 (43.4%) who underwent MT and 39,326 (56.6%) who underwent PCI. The mean age of the patients included in the articles was 63.6 years ± 5.8, with an average percentage of male 66.6% ± 12.8. Only 31/51 (60.8%) papers specified a prevalence of preexisting chronic kidney disease (CKD) from 2.0% to 50.3%. A total of 45/51 (88.2%) articles reported a diabetes prevalence ranging from 11.1% to 53.0%. 46/51 (90.2%) studies reported smoking status, with a prevalence ranging from 7.5% to 72.0%.

The meta-analysis demonstrated high heterogeneity with I^2 98.32 and Cochrane’s Q 2984.77.

The AKI incidence in the entire population was 12.4% (9,390/69,464) (confidence interval (CI95%) 10.6 to 14.3]. The incidence of AKI in the MT subgroup was significantly lower compared to PCI subgroup (8.3% [CI95% 4.7 to 11.9] vs 14.7 [CI95% 12.6 to 16.8]; p<0.05) (Figure 2a.). The difference remained significant after removing the study by Alqahtani et al.(18) (7.8% [CI95% 5.1 to 10.5] vs 14.7% [CI95% 12.6 to 16.8], respectively).

• Download figure
• Open in new tab

Figure 2. a) showed the meta-regression subgroups analysis; b) displayed the leaving-one-out meta-analysis

Sensitive analysis

Leave-one-out meta-analysis did not reveal any significant differences (Figure 2b). Furthermore, the OR remained similar after removing Cereda et al.[19] (OR 1.08 [CI95% 1.01 to 1.16]), who derived data from an RCT (DEFUSE 3), and Alqahtani et al.(18), who impacted >70T of the sample size in the MT subgroup(OR 1.08 [CI95% 1.01 to 1.16]). Funnel plots showed asymmetry, which demonstrated moderate publication bias (Figure 3).

• Download figure
• Open in new tab

Figure 3. Funnel plot

Perspective

The present study demonstrated that CSIL during endovascular procedures (MT and PCI) may significantly and independently reduce the incidence of AKI in an emergency setting, including IS and AMI.

AKI is a major problem during hospitalization, accounting for an incidence of 10-15% for all hospitalizations, which increases to 50% in ICU patients(2). The incidence of AKI in both AMI and IS is similar (7.9 for PCI vs. 5.0% for MT (19,20)). AKI may occur for several reasons, including clinical conditions, drugs, and contrast medium administration(2). The occurrence of AKI during myocardial infarction increased the chance of mortality or major cardiovascular events, with a hazard ratio of 7.9, (95% CI 2.5 to 13)(21).

The pathophysiology of AKI depends on conditions associated with its develop(2). Concerning the development of AKI during stroke or myocardial infarction treated with endovascular procedures (MT and PCI, respectively), the pathophysiology is similar, and includes intermittent hypovolemia, inflammation, and contrast medium administration(22,23). In addition, a cardiac pump deficit after AMI may play a role in the development of AKI by perturbations in the renal blood flow, leading to an imbalance in oxygen supply and demand(2). Unfortunately, no MT studies in this meta-analysis reported the ejection fraction (EF%) or heart failure score systems (e.g., Killip classification). However, the pooled absolute value of EF%, in the PCI subgroups, was only slightly inferior comparing with normality (49.1% ± 4.0)(24). Furthermore, the prevalence of Killip score >1 is inferior to that reported in the literature (11.3% vs. 33.2%) after AMI (25). In addition, cardiovascular damage is the second leading cause of death after stroke, and 28.5% of patients had LVEF < 50%, and 13–29% had systolic dysfunction at 3 months(26). This evidence suggests that although a direct comparison regarding cardiac pump deficit is not possible due to lack of data, it can be assumed that both populations are similar in this respect.

Hydration plays a key role in the prevention and treatment of AKI (Level B evidence)(13). CSIL is a standard operative procedure during MT to reduce the risk of thrombus formation inside the catheters(8). The idea behind this study is that CSIL, which has the secondary effect of fluid expander, during MT is responsible for the lower incidence of AKI compared with that after PCI in AMI. To demonstrate this, we performed a subgroup meta-analysis comparing the incidence of AKI in both the MT and PTCA data subsets (MT 8.3% [CI95% 4.7 to 11.9] vs PCI 14.7 [CI95% 12.6 to 16.8]) (Figure 2a. and 2b.). These data were similar to those extracted from the literature for IS (5.0% [95% CI, 2.1% to 8.9%])(20) and AMI (12.8% [95% CI 11.7 to 13.9%]) (15).

Although acute myocardial infarction and ischemic stroke share a clinical picture, prognosis, and effective treatment, some differences exist(5). However, the admission rates in intensive care and short-term mortality were similar between IS and AMI (88% vs. 82.3% and 15% vs. 10%, respectively), showing a similar clinical course(27–30). In this systematic review, the amount of contrast media injected intra-arterial was significantly higher during PCI comparing with MT (198.6 [CI95% 178.1 to 219.30] vs 120.5 [CI95% 99.2 to 141.7]). However, the meta-regression did not demonstrate a significant correlation between AKI and the amount of contrast medium injected (OR 1 [CI95 1-1.001], p=0.33).

Nevertheless, to balance the clinical conditions between the two diseases, meta-regression using clinical/demographic details as covariates was performed (Table 2). As hypostatized in the multivariate analysis, the absence of CSIL was associated with a significant increase in the risk of developing AKI (OR 1.08 [CI95% 1.001–1.16]; p = 0.03).

However, all MT use CSIL and, conversely, all PCI do not; therefore, a multicollinearity problem is evident. However, the results have a strong pathophysiological meaning: CSIL is a fluid expander, which is the treatment/prevention strategy for AKI(31,32).

In fact, from a scientific/clinical point of view, since hydration using saline is the standard strategy for treatment and preventing AKI, CSIL during MT may have the same effect (13,33,34). Although it is not possible to determine the exact amount of saline administered during the procedure, it can be estimated in a range of 361.8 to 1085.4 ml for a mean procedural time of 67 min and 3 catheter lines(10,11). The estimated amount of saline administered during MT is similar to that suggested by the ESUR Contrast Medium Safety Committee guidelines (800 ml for a patient weighing 80 kg)(13). This hypothesis is also indirectly supported by the lower incidence of AKI depicted between patients treated with MT and all ischemic stroke (1.6% compared with 4.2%; p = 0.0071)(35). This is even more important considering that patients treated with MT tend to have more severe clinical conditions and more contrast media administration (also intra-arterial), and therefore, a higher theoretical expected AKI incidence(6). However, the incidence is significantly lower (relative risk MT 0.4 [95% CI 0.2 to 0.8], p<0.007) (35). On the contrary, the administration of fibrinolytic therapy (rt-PA) increase the chance of having AKI (relative risk 1.9 [95% CI 1.1 to 3.3], p<0.03)(35). A potential explanation for this apparent discrepancy is the CSIL during MT. Furthermore, this lower incidence of AKI reported for MT is not observed in acute myocardial infarction treated or not treated endovascularly (with PCI 8.6% versus without PCI 10.9%, p=0.12 and OR 0.79 [95% CI 0.60 to 1.05], p=0.10)(36). Furthermore, after propensity score one-on-one match, the incidence between PCI and no-PCI patients is similar (8.6% vs 10.9%, OR 0.77 [95% CI, 0.56–1.06] p=0.12)(36).

The potential reduction in the incidence of AKI during PCI for AMI using a simple method such as CSIL has relevant clinical implications. In fact, the occurrence of AKI during AMI increases the chance of death with an OR of 12.52 (95% CI 9.29–16.86)(37). In addition, for nine cases of AKI successfully avoided during AMI treated with PCI, one death could be prevent(37).

This study has several limitations. First, although the results were consistent with the pathophysiologywasevention, and treatment theof AKI, multicollinearity remains a potential source of bia.sAlthough, a direct comparison regarding cardiac pump deficits were not possible, although, literature permits the assumption of clinical similarities although the results were consistent with the pathophysiology, prevention, and treatment of AKI, multicollinearity remains a potential source of bias. Third, ecological bias is an unresolved issue in meta-regressions. Fourth, we did not re-perform the meta-analysis by Lun et al.(15) concerning the incidence of AKI in PCI; however, the study had a large sample size (>1.2 million patients) and updated the data (published in 2021).

In conclusion, CSIL significantly reduced the incidence of AKI during endovascular procedures in the emergent setting including ischemic stroke and acute myocardial infarction. Due to the limitations of the present review, a randomized controlled trial that compared the efficacy of CSIL in PCI is mandatory owing to its clinical implications.