Screening tools for early neuropsychological impairment after aneurysmal subarachnoid hemorrhage

Aneurysmal subarachnoid hemorrhage (aSAH) is a life-threatening disease with short- and mid-term mortality of roughly 40% [1]. In recent years, survival rates have increased due to improvements in risk factor control, early diagnosis, and critical care management, as well as in surgical and endovascular prevention of re-bleeding [2]. While the number of aSAH survivors is slowly rising, the assessment of patients with non-fatal outcomes is becoming increasingly relevant [3].

During the hospital period following aSAH, the outcome of survivors is often evaluated only by functional grading scales such as the modified Ranking Scale (mRS) [4] or Glasgow Outcome Scale (GOS) [5]. Despite their advantages in terms of standardizing the outcome assessment, some criticism has been raised regarding their insensitivity to neuropsychological deficits (NPDs) [6‐8]. In fact, NPDs are the most common form of disabilities after aSAH; nearly half of the independent patients suffer from these impairments, causing difficulties in activities of daily living (ADLs) and return to premorbid work [9‐11]. It has been shown that early initiation of appropriate rehabilitation after aSAH can decrease the rates of disabilities [12, 13]. Nevertheless, early-phase screening tools that include the neuropsychological and psycho-social aspects of patient outcome are rarely conducted in neurosurgical practice or clinical trials [14].

In a recent study, Haug Nordenmark et al. [15] concluded that the majority of early NPDs after aSAH are missed by the mRS, hence additional assessments should be included in neurosurgical practice. We aimed to determine whether two practical screening tools for NPDs can identify early neuropsychological impairments, even among aSAH patients categorized with favorable outcome that might be discharged instead of receiving inpatient rehabilitation. If a screening tool detected NPDs – even in favorable outcome patients – it could be considered for implementation in a routine discharge assessment. Diagnosing NPDs would enable initiating appropriate therapy, thereby decreasing long-term morbidity of aSAH.

Of the 170 aSAH patients, 42 (24.7%) performed at least one out of the two specific NPD tests during hospitalization, and thus were included in our analysis. The specific proportions for the MoCA and ERFC were 18.2% (n = 31) and 9.4% (n = 16), respectively. On average, the patients included in the analysis showed milder aSAHs (p = 0.002), hospitalization periods that were six days shorter (p = 0.011), and more favorable hospital discharge status (p = 0.024) compared to aSAH patients who were evaluated only by the mRS (Table 1). NPD screenings were performed on average sixteen (interquartile range (IQR) = 13–21) days after aSAH whereas median difference to mRS assessments was two (IQR = 0–5) days. None of the analyzed patients had excellent functional outcome (mRS = 0) at discharge.

The results of this preliminary and retrospective study suggest that the MoCA and ERFC are valuable instruments for the early detection of NPDs in aSAH patients that are in relatively good condition and might be considered for home discharge instead of inpatient rehabilitation. We found NPDs in more than half of the patients that were categorized with favorable outcome according to the mRS (using the described cut-off values for cognitive impairment). NPDs and mRS scores showed no correlation, suggesting that functional and neuropsychological outcomes should be assessed independently. As discharge status and the perceived disabilities at discharge are key indicators for the choice of proper rehabilitation, the simple outcome screening tools examined here would likely help to recognize those aSAH patients who would benefit from a more detailed neuropsychological workup and early neuropsychological rehabilitation.

The favorable outcome status has generally included mRS categories 0–2. However, in order to exclude patients with slight disabilities, a more strict definition has been proposed for this status, which only includes mRS categories of 0 and 1 [33]. Our study shows that the mRS is unable to detect neuropsychological impairments, regardless of the cut-off range. In fact, according to the MoCA and ERFC, the proportions of NPDs were even higher (66.7% and 77.8%, respectively) in aSAH patients with good mRS outcome (mRS = 1). This finding suggests that even when the function of aSAH patients appears similar to the pre-ictal level, simple but validated neuropsychological screening tools can reveal impairments that might otherwise remain unnoticed. Prior works from Haug Nordenmark et al. [15] elucidated a significant mismatch between an – apparently – favorable outcome on the mRS and severely impaired global cognitive function in the acute phase (close to discharge from the hospital). It is possible that with ongoing recovery over time the cognitive scores and disability grading scales align better, however. At three months after the ictus, Wong et al. [27] demonstrated a moderate negative correlation between the mRS and Chinese version of the MoCA (−0.413, p < 0.001). One drawback of both our current and the previous study from Haug Nordenmark et al. [15] are the limited sample sizes, which cannot exclude the possibility that the analyses were underpowered to detect a significant correlation. Still, the impressive contrast between relatively mild functional impairment and severely reduced cognitive abilities outlined in both studies leave little room for doubt that the mRS does not qualify to judge upon the need for neuropsychological rehabilitation after hospital discharge.

It has been 25 years since Hütter and Gilsbach highlighted that six months post-aSAH, neuropsychological deficits were frequent even among SAH patients that seemed to fare well. [6] Since then, similar findings have been reported by several other research groups [11, 34, 35]. However, we found only one study that evaluated the incidence of NPDs in patients with favorable functional outcome at the early stages after aSAH. Specifically, Haug Nordenmark et al. [15] reported that 57% of all discharged aSAH patients had a poor cognitive outcome. Similar to our results, nearly half of the cases were still graded with favorable outcome (mRS 0—2). Their conclusion was that in addition to the mRS, another neuropsychological screening tool should be used to identify early NPDs after aSAH. However, the authors did not suggest any specific, short and practical screening tool – which is what we have provided in the present paper.

One of our screening tools, the MoCA, has also been considered by the NIH/NINDS CDE as a first-line neuropsychological screening tool among aSAH patients [22]. The sensitivity for milder NPDs is reportedly high, and the tool has therefore been graded as superior to other commonly used cognitive screening tools, such as MMSE [19, 26]. In fact, in the recent study of 337 aSAH patients, Eagles et al. [35] concluded that although MMSE scores varied in favorable outcome patients (mRS 0—2) 12 weeks after aSAH, more sensitive variables, such as the MoCA, should be used to detect milder NPDs among these patients. Similarly, three studies by Wong et al. have investigated the relationship between the mRS and MoCA [20, 27, 36]. The first study revealed a low association between the MoCA (measured at a subacute phase (2—4 weeks) after aSAH) and unfavorable functional outcome (mRS 3—5) at one year [36]. In the second study, a moderate negative correlation was found between the Chinese version of the MoCA and mRS three months after aSAH [27]. In the third study, only an excellent outcome (mRS = 0) was associated with the MoCA at one year [20]. Similar to our findings, these results emphasize that the MoCA qualifies as a rational tool to screen the neuropsychological function in aSAH patients. Moreover, our ERFC findings are in line with those of the MoCA. However, as the literature behind the ERFC is limited, we believe that more data is needed to confirm its ability to reliably diagnose SAH-associated NPDs.

Neuropsychological impairments were frequently identified among aSAH patients that were categorized with favorable outcome according to the mRS. The current work highlights this previously described phenomenon in a contemporary series of aSAH patients managed according to modern guidelines. Moreover, the data show that by employing a reasonably short and inexpensive neuropsychological screening tool, a substantial proportion of patients is identified who might benefit from early neuropsychological rehabilitation, even though they appear to fare well. Further studies are necessary to evaluate the actual long-term benefit of such an early screening for the patient.