Introduction

Patients with aneurysmal subarachnoid hemorrhage (SAH) may develop transient cardiopulmonary dysfunctions, such as Takotsubo cardiomyopathy (TCM) or neurogenic pulmonary edema, although their occurances are relatively rare [4, 6, 7, 11, 18]. A massive release of catecholamines into the systemic circulation after aneurysmal rupture has been considered responsible for SAH-induced TCM [12]. Thus far, no guidelines have been published for the optimal treatment of SAH patients complicated by TCM [2]. Because of their unstable hemodynamics, clipping under general anesthesia is considered to carry substantial risk, and endovascular coiling has been favored as the first-line treatment for obliterating the ruptured aneurysms [8, 13, 14]. Nevertheless, there may be instances in which clipping is preferable, such as in patients with concomitant massive intracerebral hemorrhage (ICH) [5]. There have been limited data in the literature on the feasibility and safety of clipping and coiling in SAH patients with TCM, as well as on their therapeutic outcomes [3]. Therefore, the objectives of this study were as follows: (1) to evaluate the feasibility and safety of clipping and coiling performed acutely in SAH patients with TCM and (2) to document the therapeutic outcomes and identify possible prognostic factors in that population.

Materials and methods

Patients

This was a retrospective analysis of prospectively acquired data: in January 2008, we developed an institutional protocol to perform transthoracic echocardiography (TTE) on patients with aneurysmal SAH within 24 h of admission. Measurements of blood biomarkers for cardiac injury were also performed prospectively. Clinical, echocardiographic and laboratory data of patients with aneurysmal SAH admitted between January 2008 and December 2014 were retrieved from our institutional database and reviewed.

Endovascular coiling has been the first-line treatment in our institution for aneurysmal SAH patients with TCM. However, in patients with a ruptured middle cerebral artery (MCA) aneurysm and/or in those with a massive ICH who require hematoma evaluation and decompressive craniectomy, clipping has been preferred to coiling for the obliteration of a ruptured aneurysm. The decision to perform either coiling or clipping in SAH patients with TCM was made jointly by a team comprising of endovascular specialists, vascular neurosurgeons, cardiologists and anesthesiologists. Patients were brought to either an operating room or angiographic suite after they had been stabilized hemodynamically. For patients selected for coiling, treatment was frequently started 6 h after onset to mitigate angiography-related aneurysmal rebleedings [9].

The SAH grading system of the World Federation of Neurosurgical Societies (WFNS) was used for analysis [16], with patients graded at the time of admission. Patients with a grade IV or V SAH were sedated and intubated in the emergency department (ED) and were managed with a ventilator in the neurosurgical intensive care unit until they were brought to an angiographic suite or an operating theater. In our institution, we made maximum efforts to perform procedures to obliterate the ruptured aneurysm in patients with grade V SAH within 24 h of symptom onset. Exceptions were made for those with prolonged/unwitnessed cardiac arrest, those with bilaterally fixed and dilated pupils, those with radiographic evidence of irreversible brain injury and those whose surrogates refused treatment.

For patients with persistent hypotension, continuous intravenous (IV) infusion of norepinephrine was used with or without dopamine to maintain the systemic circulation. For maintenance of general anesthesia, we employed a combination of sevoflurane, propofol and remifentanil. Postoperative management included the placement of a lumbar catheter to evacuate the subarachnoid clot (particularly in patients who underwent coiling) and the IV use of intravenous fasudil hydrochloride (Asahikasei Pharma, Miyazaki, Japan), low-molecular-weight dextran and albumin to prevent vasospasm. Patients treated by coiling also received oral aspirin as an antithrombotic. Patients underwent brain computed tomography (CT) routinely on the 12–14th day after obliteration of an aneurysm to evaluate the presence of delayed cerebral ischemia (DCI) associated with vasospasm. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Approval for this study was given by our institutional research committee. For this type of (i.e., retrospective) study, formal consent is not required from each patient.

Echocardiographic evaluations

Maximum efforts were made to obtain TTE before treatment of a ruptured aneurysm to ensure that therapeutic decisions could be based on the cardiac status of each patient. For acquisition of TTE, we used a General Electric Vivid 7 (GE Healthcare, Tokyo, Japan). All TTE examinations were performed by experienced ultrasound technicians based on the guidelines of the American Society of Echocardiography [9], and echocardiographic videos were reviewed by board-certified cardiologists. The diagnosis of TCM, defined as reginal cardiac wall motion abnormality extending beyond a single epicardial vascular distribution, was based on the revised Mayo Clinic Criteria [17].

Cardiac biomarkers

The following biomarkers of cardiac injury were measured: norepinephrine, troponin I, B-type natriuretic peptide, white blood cell count and blood glucose. In all patients with aneurysmal SAH, blood samples were collected and sent to the laboratory within 24 h of admission. Maximum efforts were made to collect blood for the measurement of norepinephrine levels before intravenous catecholamines were administered.

Perioperative adverse events and mortality

To evaluate the feasibility and safety of each treatment modality, the frequency of perioperative adverse events was compared between patients treated by clipping and those treated by coiling. The following variables were compared: the incidences of refractory hypotension during treatment, the necessity for the use of inotropes and the development of postoperative heart failure, postoperative pulmonary edema, postoperative acute renal failure and DCI. Demographic variables and therapeutic outcomes were also compared. For patients who died within 90 days of admission, the cause of death was investigated by a neurosurgeon who was not directly involved in patient care.

Outcome evaluation

Patient outcomes were assessed at 90 days by a same neurosurgeon responsible for assessing the perioperative morbidity and mortality, using the modified Rankin Scale (mRS). An mRS of 0–2 at 90 days defined a favorable outcome. The SAH patients with TCM were dichotomized based on their 90-day mRS scores into favorable and unfavorable groups, and the demographic variables, physiologic variables and laboratory data were compared between the two groups. Furthermore, multivariate logistic regression analysis was performed to identify clinical variables predictive of favorable outcomes. The variables evaluated were as follows: age, sex (males), posterior circulation aneurysm, WFNS grade V SAH, coiling as a treatment for a ruptured aneurysm, the presence of concomitant pulmonary edema, left ventricular ejection fraction (LEVF) <40 % (an indication of moderate to severe heart failure) [19] and DCI.

Relationship between patient age and outcomes

Receiver operating characteristics (ROC) curve analysis was conducted to calculate the threshold age predictive of favorable outcomes. From the ROC curve, we derived the optimal threshold age for distinguishing between patients with and without favorable outcomes by seeking the best trade-off between highest possible sensitivities and specificities of the threshold values using the area under the curve (AUC). Furthermore, the 391 aneurysmal SAH patients were quadrichotomized on the basis of the threshold age and sex, and the incidences of TCM in the four groups were compared.

Statistical analysis

The Fisher’s exact test was used to compare group differences in categorical variables, and the Student’s t-test was used to compare group differences in numerical variables. Multivariate logistic regression and ROC curve analyses were performed using JMP software (SAS Institute, Cary, NC, USA). Numerical data are expressed as the mean ± standard deviation. A p-value < 0.05 was considered statistically significant.

Results

Demographics

During the 7-year study period, 391 consecutive aneurysmal SAH patients were admitted to our institution and were treated for ruptured saccular aneurysm (n = 350) or dissection (n = 41). Of these, 129 were males, with a mean age of 54.9 ± 11.4 years, and 262 were females, with a mean age of 62.9 ± 13.2 years. The distribution of admission WFNS SAH grade was shown in Fig. 1a. The mean interval between symptom onset and ED arrival was 2.8 ± 3.5 h, and the mean ED arrival to surgery (clipping/coiling) interval was 11.1 ± 8.7 h. Among the 391 SAH patients, 30 (7.7 %) were diagnosed with TCM by cardiologists based on their TTE findings. The male-to-female ratio was 11:19, with a mean age of 52.6 ± 13.3 years for males and 67.1 ± 12.9 years for females; the males were significantly younger than females (p = 0.007). None had a prior history of acute coronary syndrome, and 29 (97 %) of the 30 patients underwent treatment of the ruptured aneurysm/dissection within 24 h of symptom onset. The distribution of the WFNS SAH scale showed that 16 (53 %) presented with WFNS grade V SAH (Fig. 1b). Eleven patients (37 %) underwent clipping, and the other 19 patients were treated by coiling.

Fig. 1
figure 1

a Bar graph showing the distribution of the World Federation of Neurosurgical Societies (WFNS) grades at admission in 391 consecutive aneurysmal subarachnoid hemorrhage (SAH) patients. b Bar graph showing the distribution of the WFNS grade at admission in the 30 SAH patients with Takotsubo cardiomyopathy (TCM). c Bar graph showing the distribution of the 90-day modified Rankin Scale scores in the 30 SAH patients with TCM

Full size image

Comparison of perioperative adverse events between lipping and coiling

Comparison between clipped (n = 11) and coiled patients (n = 19) showed that the frequency of ruptured MCA aneurysms was significantly higher in patients treated by clipping (55 % compared with 5 %, p = 0.005) (Table 1). There were no significant intergroup differences in other demographic variables (Table 1). Similarly, there were no significant differences in the frequencies of intraoperative adverse events or outcomes between the two groups (Table 1). Although intraoperative bleeding occurred in one patient in both groups, intractable hypotension refractory to vasopressors did not occur in either group. One patient treated by coiling developed both heart failure and pulmonary edema postoperatively, but the complication was temporary and did not affect the outcome. None of the 30 patients with TCM required the use of balloon-pump counterpulsation. Similarly, no patients developed acute kidney failure postoperatively.

Table 1 Comparison between clipped and coiled patients
Full size table

Cause of death

The causes of death in six patients who died within 90 days of admission are summarized in Table 2. In four of the six patients, causes of death were treatment related (mostly post-procedural rebleeding), and another two died from DCI. Heart failure attributable to TCM was not considered responsible for their death in any of the patients.

Table 2 Causes of death within 90 days of onset
Full size table

Comparison of outcomes between the favorable and unfavorable groups

The 90-day mRS scores in the 30 patients with TCM are illustrated in Fig. 1c, showing that 13 (43 %) had 90-day mRS scores of 0–2 (i.e., favorable), and the other 17 had 90-mRS scores of 4–6 (i.e., unfavorable). Comparison of the demographic and physiologic variables between the two groups is summarized in Table 3. The favorable group was significantly younger (53.7 ± 13.8 years compared with 67.9 ± 12.4 years, p = 0.006) and tended to have a lower frequency of DCI (8 % compared with 41 %, p = 0.09). There were no significant intergroup differences in other demographic variables, including sex, percentage with grade V SAH, clipping-to-coiling ratio, the frequency of posterior circulation aneurysms and concomitant pulmonary edema. Similarly, there were no significant intergroup differences in the physiologic variables such as admission systolic blood pressure and LEVF.

Table 3 Comparison of demographic and physiologic variables between patients with favorable and unfavorable outcomes
Full size table

Echocardiographic/laboratory evaluation

Among the 30 SAH patients with TCM, TTE could be obtained before clipping/coiling in 26 (87 %). Their LVEF ranged from 22 to 65 % (mean: 43.7 % ± 9.9 %). According to the Acute Decompensated Heart Failure Registry Criteria [18], 2 patients had severely decreased LVEF (<25 %), 6 had moderately decreased LVEF (25–39 %), 19 had mildly decreased LVEF (40–54 %), and 3 patients had a normal LVEF (≥55 %). In total, 13 patients (43 %) required temporary use of inotropes perioperatively for hypotension. In addition, there were no significant differences in the cardiac biomarker levels between the favorable and unfavorable group (Table 4).

Table 4 Comparison of cardiac biomarker levels between patients with favorable and unfavorable outcomes
Full size table

Multivariate logistic regression analysis

Age correlated with outcomes in the logistic regression analysis (odds ratio, 0.896; 95 % confidence interval, 0.812–0.989, p = 0.03) (Table 5). There were no significant correlations between other variables and outcomes, including DCI (Table 5).

Table 5 Multivariate logistic regression analysis to identify variables associated with favorable outcomes
Full size table

ROC curve analysis

ROC curve analysis showed that the threshold age for favorable outcomes was 65 years, with an AUC value of 0.81 (Fig. 2). Subsequently, the 391 SAH patients were quadrichotomized (males vs. females, ≤65 years vs. >65 years of age). The incidences of TCM were 9 % (9/105) in males ≤65 years, 8 % (2/24) in males >65 years and 9 % (11/124) in females ≤65 years, and 6 % (8/138) in females >65 years (Fig. 3). There was no significant intergroup difference in the frequency of TCM between any of the two groups.

Fig. 2
figure 2

The optimal threshold age to distinguish between patients with and without favorable outcomes was derived from the receiver operating characteristics curve. The threshold age was 65.0 years, with an area under the curve value of 0.81

Full size image
Fig. 3
figure 3

The 391 SAH patients were quadrichotomized on the basis of sex and age (≤65 years vs. >65 years of age). There were no significant differences in the frequencies of SAH-induced TCM between any of the two groups

Full size image

Discussion

In the present study, 63 % of the 30 patients with SAH-induced TCM were treated by coiling, and the remaining 37 %, most of whom harbored either a ruptured MCA aneurysm and/or a concomitant massive ICH, were treated by clipping. There were few perioperative cardiopulmonary complications in those who underwent clipping (Table 1). Endovascular coiling was also performed safely: despite the risk of acute renal failure in hypotensive patients in whom vasopressors such as norepinephrine are used, none of the 19 patients treated by coiling developed that complication (Table 1). However, procedure-related complications in coiled patients were relatively high, considering that 3/19 (16 %) developed fatal complications (post-procedural rebleeding in 2 and treatment-related cerebellar infarction in 1) (Table 2). In retrospect, some of those patients, particularly those with irregularly shaped aneurysms considered unfit for coiling, might have been better treated by clipping. Although coiling has been considered the first-line treatment in our institution, therapeutic decisions may better be made more flexibly and on a case-by-case basis in the absence of severe cardiac dysfunction (LVEF <25 %). The incidence of SAH-induced TCM in our cohort, i.e., 391 consecutive patients who underwent TTE in a prospective fashion shortly after admission, was 7.7 %. According to a recent retrospective study, 31 (3.7 %) of 828 patients with aneurysmal SAH evaluated by TTE exhibited reduced LVEF with regional wall motion abnormality [1]. The higher incidence of TCM in this study may have been due to the relatively high percentage of poor-grade SAH patients: indeed, >40 % presented with either WFNS grade IV or IV SAH (Fig. 1a).

Regarding the outcomes, age seems to be the only prognosticator, as shown in the two-group comparison (Table 3) as well as in the logistic regression analysis (Table 5). Younger age has long been known to be associated with favorable outcomes in patients with SAH regardless of the presence of TCM [2, 10, 16], which is consistent with the current results. We also showed that a cutoff age of 65 years produced an acceptable AUC value of 0.81 (Fig. 2) and could serve as a crude predictor of outcomes in SAH patients with TCM. Contrary to the previous finding that older females are more likely to develop TCM [12], the incidences may not have been affected significantly by either sex or age in this study (Fig. 3). The incidence of DCI tended to be higher in the unfavorable group (Table 3), with it being likely that DCI is causally related to poor mRS scores in afflicted patients. However, DCI was not correlated with unfavorable outcomes by the logistic regression analysis (Table 5), probably because not a few severe adverse events unrelated to DCI had occurred perioperatively, such as fatal post-procedural rebleedings (Table 2).

Interestingly, neither a lower LEVF nor the presence of concomitant pulmonary edema correlated with unfavorable outcomes (Table 5). Similarly, the lack of any difference in cardiac biomarker levels between the favorable and unfavorable groups (Table 4) indicates that the degree of cardiac injury may not be a crucial factor in determining the outcomes of SAH patients with TCM. Analyzing the causes of deaths (Table 2) revealed that patients tended to die from treatment-related complications rather than heart failure. Thus, aggressive therapeutic intervention in conjunction with cardiologists and anesthesiologists may be warranted in patients with severely impaired cardiac function because the impairment is almost invariably transient. For patients with hypotension refractory to inotropes, several authors, whose studies included patients with an LVEF as low as 10 %, reported the efficacy of balloon pump counterpulsation [3, 15]. Fortunately, inotropic support was sufficient to ameliorate hypotension in our cohort, and no patient required balloon pump counterpulsation. Compared with the aforementioned studies [3], the degree of systolic dysfunction might have been milder in this study: only 2 (7 %) of the 30 patients sustained severe systolic dysfunction (i.e., LVEF < 25 %), and inotropes were required perioperatively in less than half of patients (43 %).

There are several limitations to this retrospective study. First, therapeutic decisions were not always made on the basis of TTE findings: in four patients (13 %) with massive ICH and impending brain herniation, TTE was obtained after treatment (clipping), which had proceeded without cardiologic consultation, possibly suggesting that such a consultation is important but not absolutely necessary. It is also possible that the surgery might have influenced cardiac function and the reported LEVF value in those patients. Second, the follow-up period of 90 days might have been too short to assess the outcomes, which might have improved in some patients after longer follow-up periods. Third, the number of patients was inevitably small: given that SAH-induced TCM has a relatively low incidence, making it difficult to accumulate an adequate sample size, it may be important to establish national/international registries. Finally, our study objectives were to confirm the safety and feasibility of clipping and coiling performed within 24 h of symptom onset in SAH patients with TCM, and we are not in a position to deny the usefulness of delayed treatment in that population. Despite these limitations, this study may be unique and valuable as one of the first case series to report the feasibility and safety of the two gold-standard treatment modalities for aneurysm obliteration as well as the therapeutic outcomes in SAH patients complicated by TCM.

Conclusions

In the present study, 63 % of the 30 patients with SAH-induced TCM were treated by coiling, and the remaining 37 % were treated by clipping. Neither clipped nor coiled patients developed serious perioperative cardiopulmonary complications, but coiled patients had a relatively higher incidence of fatal procedure-related complications. Clipping is a safe and effective treatment modality in most SAH patients complicated by TCM, and treatment selection may be made on a case-by-case basis. The lack of correlation between the degree of cardiac dysfunction and therapeutic outcomes indicates that aggressive intervention is justified in patients with severely impaired cardiac function.