What's new?

Mechanical thrombectomy (MT) is the gold standard of treatment in acute ischemic stroke (AIS) caused by large vessel occlusion. The procedure requires the use of iodinated contrast, and therefore entails a risk of contrast‑associated nephropathy. Chronic kidney disease (CKD) is a common comorbidity in stroke patients and is associated with worse prognosis. The number of studies analyzing the outcomes of MT in AIS patients with CKD is limited, especially concerning long‑term follow‑up. We showed that 90- and 365‑day outcomes of MT are worse in CKD patients, especially those with more severely impaired kidney function, but at the same time CKD was not found to be independently associated with poor prognosis in this group. Therefore, CKD alone should not be considered a contraindication for the procedure in otherwise eligible patients.

Introduction

Chronic kidney disease (CKD) is defined as kidney structural or functional abnormalities that impact the patient’s health and last for a minimum of 3 months. It may present in laboratory testing as glomerular filtration rate (GFR) below 60 ml/min/1.73 m², albuminuria above 30 mg/24 hours, or other markers of kidney damage.¹ CKD is a common disease, affecting 8%–16% of the world’s population.² Globally, its most common cause is diabetes mellitus, although it may be associated with various other factors, mostly acquired, but sometimes also congenital.³

CKD significantly increases the risk of mortality due to cardiovascular events.⁴ Acute ischemic stroke (AIS) and CKD share many risk factors, but studies show that CKD itself is an independent risk factor of AIS.^5,6 At the same time, AIS outcomes are significantly poorer in patients with CKD.^7,8

Mechanical thrombectomy (MT) is the gold standard of treatment in AIS caused by large‑vessel occlusion.⁹ Both the procedure and the qualification for MT require the use of iodinated contrast, therefore, there is a risk of treatment‑associated contrast‑induced nephropathy (CIN), a condition that mostly occurs in patients with pre‑existing CKD.¹⁰ Acute kidney injury (AKI) due to various causes, including CIN, is seen in around 7% of MT‑treated AIS patients, and CKD is one of the factors associated with this complication.¹¹ Nevertheless, the incidence of end‑stage kidney failure requiring dialysis after the procedure is low.^12,13 According to Polish guidelines¹⁴ there is no need to routinely assess serum creatinine concentration (SCr) before qualification for MT in patients without a CKD history, as the risk of CIN in such cases is negligible.¹⁴

A significant number of patients with CKD receives treatment with MT (in 1 meta‑analysis they represented 28% of all MT‑treated patients).¹⁵ At the same time, not many studies assess their prognosis, and, to the best of our knowledge, none analyzed outcomes for longer than 6 months. Therefore, the aim of this study was to assess the short- and long‑term outcomes of MT (up to 1‑year of follow‑up) in CKD patients.

Patients and methods

The study included all AIS patients treated with MT at the Comprehensive Stroke Center of the University Hospital in Kraków, Poland, between 2019 and 2021. We performed a retrospective analysis of data prospectively collected for the purpose of a study entitled Identification and clinical validation of biomarkers for long‑term outcome after cerebral ischemia (IBioStroke) (https://www.neuron‑eranet.eu/wp‑content/uploads/iBioStroke.pdf).

We divided the participants into 2 groups, that is, the CKD group including patients with a known history of CKD confirmed in their medical records or with CKD diagnosed during hospitalization based on kidney function test results, if the best estimated GFR (eGFR), assessed using the CKD Epidemiology Collaboration equation, obtained during hospital stay was below 60 ml/min/1.73 m², and the control group comprising the remaining patients, without a history of CKD. The CKD group was subsequently divided into 2 subgroups based on the CKD stage: 1–3 (eGFR ≥30 ml/min/1.73 m²) and 4–5 (eGFR <⁠30 ml/min/1.73 m²).

We gathered information on the patient sex, age, and cardiovascular risk factors known on admission. The severity of stroke was defined as neurological deficits on admission assessed with the National Institutes of Health Stroke Scale (NIHSS) and lesion (penumbra and infarct) volumes assessed with computed tomography perfusion postprocessing analysis with iRAPID package (iSchemaView, Menlo Park, California, United States). Data on intravenous thrombolysis (IVT) preceding MT were collected. We analyzed the course of the procedure including the time from stroke onset to groin puncture, the rate of successful recanalization, defined as modified Treatment In Cerebral Infarction score of 2b–3, and the occurrence of complications in the form of secondary intracerebral hemorrhage (ICH) and AKI, defined as an increase in a baseline SCr level of at least 26 µmol/l within 48 hours or at least 1.5 times within 7 days.

Short‑term outcomes were defined as in‑hospital mortality, NIHSS score at discharge, and modified Rankin Scale (mRS) result at discharge, with mRS score equal to or below 2 considered a good functional outcome. Long‑term outcomes (mortality and mRS score) were assessed 90 days and 365 days after stroke, either during scheduled visits in our Center’s outpatient clinic or through telephone interviews with patients or their caretakers.

Results

Our study included 593 MT‑treated AIS patients of which 59 (9.9%) had a diagnosis of CKD. The patients with CKD were older than the controls, as their median (IQR) age was 78 (71–86) vs 70 (61–78) years (P <⁠0.001), and they were more commonly women (66.1% vs 45.5%; P = 0.004).

The CKD patients had more severe neurological deficits on admission than the controls (median NIHSS score of 17 [IQR, 13–20] vs 15 [IQR, 10–19]; P = 0.018) but infarct and penumbra volumes (available in 521 patients) were comparable. The CKD patients were less often treated with IVT (40.7% vs 58.8%; P = 0.009). There were no differences between the groups in the time from stroke onset to groin puncture, successful recanalization rate, and the percentage of patients developing secondary ICH (data available in 566 patients). AKI diagnosed according to the Kidney Disease: Improving Global Outcomes recommendations¹ was more common in the CKD patients (25.4% vs 11.2%; P = 0.004).

The patients with CKD did not have much higher in‑hospital mortality (22% vs 12.7%; P = 0.069), and the percentage of individuals achieving good functional outcome and median NIHSS score at discharge also did not differ significantly between the groups. In the group of 59 CKD patients, there were no significant differences in in‑hospital mortality and rate of good functional outcomes at discharge between those treated with IVT+MT and MT alone.

Mortality rate was higher and the percentage of good functional outcomes lower in the CKD group, both in 90‑day observation (39.7% vs 21.2%; P = 0.002 and 34.5% vs 59.2%; P <⁠0.001, respectively) and in 365‑day follow‑up (49.1% vs 29%; P = 0.003 and 36.8% vs 62.8%; P <⁠0.001, respectively). A detailed group comparison is showed in Table 1.

The patients with more severe CKD (stages 4–5) more often developed AKI than those with CKD stages 1–3 (70% vs 16.3%; P = 0.001), and had higher mortality during hospitalization as well as in 90- and 365‑day follow‑up (60% vs 14.3%; P = 0.005; 80% vs 31.3%; P = 0.006; and 88.9% vs 41.7%; P = 0.012, respectively). A comparison of patients with different CKD stages is shown in Table 2.

When only the subgroup of patients with CKD stage 1–3 was compared with the controls (patients without CKD), the only considerable difference concerning outcomes was seen in 90‑day and 365‑day good functional outcome rate, which was lower in the CKD group (39.3% vs 59.2%; P = 0.01 and 41.7% vs 62.8%; P = 0.005, respectively). This comparison is presented in Table 3.

The univariable logistic regression analysis showed that any stage of CKD was associated with 90‑day and 365‑day mortality and good functional outcome. CKD was not associated with in‑hospital outcomes. When only the stages 4–5 of CKD were considered in the univariable analysis, they were associated with in‑hospital, 90‑day, and 365‑day mortality and functional outcomes. The results of the univariable analyses are presented in Table 4.

In the multivariable analyses, CKD was not independently associated with short- and long‑term mortality and functional outcomes of MT. The results of the multivariable analyses are presented in Table 5.

Discussion

The results of our study show that long‑term MT outcomes are worse in patients with CKD, but CKD is not an independent risk factor for poor prognosis following the procedure. The level of kidney function impairment seems to play an important role, as short- and long‑term mortality were especially high among patients with CKD stage 4–5. At the same time, when only the patients with CKD stage 1–3 were compared to controls, no differences in short- and long‑term mortality were found, and the only significant difference was found for good functional outcome rate in 90- and 365‑day follow‑up. The lower rate of good functional outcomes in this group may be associated with presumably worse prestroke mRS among CKD patients, but, unfortunately, we did not have access to this information in our group, which is a significant limitation of our analysis. Other limitations include a small number of patients with CKD stage 4–5 (n = 10), making it impossible to thoroughly analyze the reasons for and factors associated with especially poor prognosis in this group.

CKD is common among patients qualified for stroke reperfusion therapies, with 1 meta‑analysis showing its pooled incidence of 43% for IVT and 28% for MT.¹⁵ In our group, the incidence of CKD was lower (9.9%), but still significant. Those high numbers are expected with CKD being a risk factor for AIS. Reasons for increased AIS risk in patients with CKD most likely include a combination of shared cardiovascular risk factors and direct consequences of renal insufficiency, such as vascular calcification, acceleration of atherosclerosis, uremic toxin effects, hypercoagulation, chronic inflammation, or oxidative stress.^5,16 At the same time, studies consistently point out toward worse prognosis of reperfusion therapies in CKD patients.

Osman et al¹⁷ showed that in MT‑treated patients CKD is associated with higher in‑hospital mortality and poorer functional outcomes at discharge. In a study by Fandler‑Höfler et al,¹⁸ impaired kidney function on admission significantly predicted worse functional outcomes in 3‑month observation in univariable analyses, but those associations lost significance in multivariable analyses. Studies by Laible et al¹⁹ and Pan et al²⁰ showed that low GFR was independently associated with higher mortality but not poor functional outcome in 3‑month follow‑up. These results contrasted with those of Xiao et al,²¹ where renal impairment was independently associated with 3‑month functional outcome but not mortality. Sutherland et al²² showed an increase of both mortality and poor functional outcome in 3‑month follow‑up in CKD patients. Sun et al²³ found CKD to be an independent factor associated with a lack of functional independence 90 days after stroke despite successful recanalization. A meta‑analysis by Rajesh et al,¹⁵ including 6 studies and a total of 33 661 patients with anterior circulation AIS undergoing MT, showed that CKD was significantly associated with poor functional outcome 90 days after the procedure. Park et al²⁴ demonstrated higher 6‑month mortality in patients with GFR lower or higher than normal.

Other studies also demonstrated worse outcomes of MT in patients with advanced CKD. Chen et al²⁵ reported that MT outcomes are particularly poor among end‑stage CKD patients, with higher rates of symptomatic ICH and 90‑day mortality, which was particularly high in dialyzed patients with significant prestroke disability. At the same time, some studies suggest that MT is still a safer reperfusion therapy for patients with end‑stage CKD than IVT.²⁶ Interestingly, another study by Laible et al²⁷ showed that renal dysfunction is associated with hemorrhagic complications after MT in patients with posterior circulation stroke, although neither the presence of CKD nor the hemorrhage affected mortality in this group.

To the best of our knowledge, this is the first study to report MT outcomes in CKD patients with follow‑up exceeding 6 months. Although, similarly as in shorter‑term observations, 365‑day outcomes are worse than in the control group, in our group CKD was not independently associated with poor prognosis.

Worse outcomes of reperfusion therapies in CKD patients may be explained, among all, by their poorer general condition, presence of significant comorbidities, altered metabolism of medications, and higher risk of infections.^2,15 Importantly, our study, as well as some previous studies,^20,25 did not show differences in successful recanalization rates, therefore it seems that CKD does not impact the MT procedure itself. Importantly, some authors point out that although patients with CKD have worse prognosis after reperfusion therapies for stroke than the general population, they still have better prognosis than CKD patients who are not treated.^15,28

An interesting observation from the multivariable analyses of our study is that although CKD was not an independent predictor of short- and long‑term mortality and poor functional outcomes, AKI was. As CKD patients (especially those with advanced CKD) are more prone to developing AKI after MT, they require especially careful postprocedural kidney function monitoring, AKI prophylaxis, and prompt treatment, in case it develops. Interventions for MT‑associated AKI include hydration therapy, avoiding nephrotoxic medications and additional contrast use, and, in the most severe cases, renal replacement therapy.²⁹ Other identified factors associated with long‑term functional outcome included, among others, age, NIHSS score on admission, and successful reperfusion, which was similar to previous studies in this field.³⁰