Neurocritical Care

, Volume 17, Issue 3, pp 343–347

Trajectory of Functional Recovery After Hospital Discharge for Subarachnoid Hemorrhage


    • Department of Neurology and NeuroscienceWeill Cornell Medical College
  • Hooman Kamel
    • Department of Neurology and NeuroscienceWeill Cornell Medical College
  • J. Claude HemphillIII
    • Department of NeurologyUniversity of California
  • Wade S. Smith
    • Department of NeurologyUniversity of California
Original Article

DOI: 10.1007/s12028-012-9772-3

Cite this article as:
Navi, B.B., Kamel, H., Claude Hemphill, J. et al. Neurocrit Care (2012) 17: 343. doi:10.1007/s12028-012-9772-3



Although there are extensive data on long-term disability after subarachnoid hemorrhage (SAH), there are few data on the trajectory of functional recovery after hospital discharge.


From October 2009 to April 2010, we prospectively followed consecutive patients with non-traumatic SAH discharged from a university hospital. Modified Rankin Scale (mRS) scores were calculated at discharge from chart review and at 6 months by standardized telephone interview. Good functional status was defined as a mRS score of 0–2, and poor status as an mRS score of 3–6. Descriptive statistics were used to assess the trajectory of functional recovery and determine the proportion of patients whose functional status improved from poor to good.


Among 52 patients with non-traumatic SAH (79 % aneurysmal) who were discharged alive, most (71 %) were discharged home. Median (IQR) mRS score was 3 (2–4) at discharge and 2 (1–2) at 6 months. Some functional recovery (any improvement in mRS score) was seen in most patients (83 %; 95 % CI, 72–93 %). Of the 28 patients with poor functional status at discharge, 16 (57 %) improved to good functional status at 6 months. All patients with Hunt–Hess grade 4 or 5 hemorrhages (n = 14) had poor functional status at discharge, but half (95 % CI, 20–80 %) recovered to a good functional status at 6 months.


Although our sample size is small, our findings suggest that a substantial proportion of patients with SAH who are disabled at discharge go on to regain functional independence within 6 months.


Subarachnoid hemorrhageFunctional outcomeRecoveryTrajectoryPredictors


Subarachnoid hemorrhage (SAH) is a devastating neurologic disease that unlike other stroke syndromes frequently occurs in relatively young, previously healthy persons [1]. Despite significant advances over the past few decades in the care of SAH patients, nearly 40 % still die before or during medical attention [2]. In addition, those who do survive are often left with significant motor and neuropsychological deficits that preclude independence and the ability to work [37].

Several clinical prediction scores utilizing data from patients’ initial neurologic status aid in the prediction of survival after SAH [8, 9]. Furthermore, multiple clinical factors measured during patients’ hospitalization correlate with survival and functional outcomes [1013]. However, patients with SAH often suffer secondary complications such as delayed cerebral ischemia and hydrocephalus, which can cause temporary neurologic handicap, delayed convalescent periods, and prolonged rehabilitation stays [14, 15]. Therefore, there is often considerable uncertainty at the time of hospital discharge about patients’ ultimate functional status, which is particularly germane since patients’ families often need to rearrange their lives around the likelihood and timing of recovery.

We sought to better delineate the extent of functional recovery after hospital discharge in patients with non-traumatic SAH. Forecasting the clinical trajectory of recovery for these patients would aid caregivers in their prognostications, thereby assisting family members with goals-of-care discussions, time commitments, and financial decisions. In addition, such information could help physicians and policy makers to determine the optimal timing and duration of rehabilitation services.


Study Setting and Design

We performed a prospective observational study involving consecutive patients with acute, non-traumatic SAH admitted to the University of California, San Francisco (UCSF) Medical Center from October 1, 2009 through March 31, 2010. This cohort included patients admitted through our emergency department or by transfer from outside hospitals, as well as patients admitted to the neurosurgical service, since at our institution these patients are jointly cared for by the neurovascular team. The on-service neurovascular physicians identified eligible patients on a daily basis as a part of a quality assurance project intended to assess the prognostic accuracy of neurologists; the details of this study have been reported in a previous publication [16]. The UCSF Committee on Human Research approved this analysis and waived the requirement for informed consent because of minimal risk to patients.

Baseline Measurements

Relevant demographic and clinical data were collected retrospectively by chart review according to the standard of published recommendations [17]. Recorded data included patients’ age; gender; race; past medical history (self-reported or documented hypertension, hyperlipidemia, atrial fibrillation, diabetes mellitus, prior stroke, coronary artery disease, congestive heart failure, and chronic renal impairment); Glasgow Coma Scale [18], Hunt–Hess grade [8], and Fisher score [19] upon admission; etiology of SAH; coiling or clipping of aneurysm if applicable; secondary complications of SAH; the presence of mechanical ventilation; length of stay; and disposition.

The diagnosis of a secondary complication of SAH—radiographic vasospasm, clinically symptomatic vasospasm, cerebral infarction, cerebral salt wasting, seizure, recurrent hemorrhage, hydrocephalus, and bacterial or fungal infection—was determined retrospectively by means of standardized definitions as outlined in the methods of a previous report [16].

Outcome Measurements

Functional outcomes were measured at hospital discharge and at 6 months by means of the modified Rankin Scale (mRS) score [20]. The mRS is a validated neurologic disability scale that has been used in several previous studies of patients with SAH [2123]. The score ranges from 0 to 6—with 0 denoting no symptoms at all and 6 signifying death. A single neurovascular fellow, who had completed an instructional course and was certified in calculating mRS scores before the start of the project, retrospectively determined patients’ mRS score at discharge by means of clinical data from chart review, including all available clinical notes from physicians, nurses, and physical and occupational therapists. The same neurovascular fellow then prospectively determined patients’ mRS score at 6 months by telephone interview by means of a validated standardized script [24]. If neurologic disability prevented a patient from participating in the 6-month interview, the patient’s closest surrogate or primary caregiver was interviewed instead.

Functional status at discharge and at 6 months was dichotomized into good and poor. Good functional status was defined as an mRS score of 0–2 (i.e., independent in activities of daily living) and poor functional status as an mRS score of 3–6 (i.e., dependent in activities of daily living or worse). Functional recovery was defined as any improvement (decrease) in the mRS score from hospital discharge to 6 months.

Statistical Analysis

Descriptive statistics were used to calculate the rates of patients’ baseline characteristics, secondary complications, functional neurologic outcomes, and trajectory of functional recovery. In an exploratory analysis, ordinal logistic regression was used to investigate the clinical factors outlined in the “Baseline measurements” section above for an association with functional recovery; these factors comprised of patient demographics, vascular co-morbidities, SAH characteristics, and secondary complications. Factors significantly associated with recovery at the univariate level (p < 0.10) were inserted into the final multivariate model and left in place regardless of statistical significance. Sensitivity analyses were performed limited to patients with aneurysmal SAH. Statistical analyses were performed by means of Stata (Version 11, Stata Corporation, College Station, TX).


During the 6-month study period, 70 patients were hospitalized with non-traumatic SAH at UCSF. Of these patients, 56 survived to hospital discharge, and 52 had 6-month outcome data (followup rate of 93 %) and were included in the final analysis. Mean age was 54 years (SD 13) and 62 % were female. Study patients’ clinical profiles were typical for patients with non-traumatic SAH (Table 1) [14, 15]. Most patients had SAH from aneurysmal rupture (n = 41, 79 %) though a few patients had alternate etiologies, such as perimesencephalic hemorrhage (10 %) and idiopathic hemorrhage (6 %). The median (IQR) length of stay was 15 (11–22) days. Discharge from the hospital was most often to home (71 %), followed by acute rehabilitation (15 %), subacute rehabilitation (8 %), and other (6 %).
Table 1

Clinical characteristics of patients with non-traumatic SAH


Total cohort (n = 52)

Mean age, years (SD)

54 (SD 13)

Women, n (%)

32 (62)

Race, n (%)


24 (46)


20 (38)


4 (8)


1 (2)


3 (6)

Vascular co-morbidities, n (%)


26 (50)


11 (21)

 Diabetes mellitus

3 (6)

 Coronary artery disease

2 (4)

 Atrial fibrillation

1 (2)

 Chronic renal impairment

0 (0)

 Congestive heart failure

0 (0)

 Prior stroke

0 (0)

Clinical grade of SAH at admission, median (IQR)

 Hunt–Hess grade

2 (2–4)

 Glasgow Coma Scale score

15 (9–15)

 Fisher score

3 (2–4)

Etiology of SAH, n (%)


41 (79)

  Anterior circulation

36 (69)

  Posterior circulation

5 (10)


25 (48)


17 (33)

 Perimesencephalic hemorrhage

5 (10)


3 (6)

 Dural arteriovenous fistula

1 (1)

 Arteriovenous malformation

1 (1)

 Amyloid angiopathy

1 (1)

Secondary SAH complications, n (%)

 Cerebral salt wasting

33 (64)

 Radiographic vasospasm

30 (58)


20 (38)

 Mechanical ventilation

18 (35)


16 (31)

 Clinical vasospasm

12 (23)

 Cerebral infarction

10 (19)


3 (6)

 Recurrent hemorrhage

1 (2)

SD standard deviation, IQR interquartile range, SAH subarachnoid hemorrhage

* One patient was clipped and coiled thus sum of aneurysmal treatments exceeds total amount of patients with aneurysms

At the time of hospital discharge, the median mRS score was 3 (IQR 2–4) and 24 patients (46 %; 95 % CI, 32–61 %) had good functional status. At 6 months, the median mRS score was 2 (IQR 1–2) and 40 patients (77 %; 95 % CI, 63–88 %) had good functional status. Some improvement in mRS score occurred in 43 (83 %; 95 % CI, 73–92 %) patients from hospital discharge to 6 months (Fig. 1). Of the 28 patients with poor functional status (mRS score 3–6) at discharge, 16 (57 %; 95 % CI, 37–76 %) improved to good functional status at 6 months. Only one patient worsened after hospital discharge and this patient died in a hospice facility after being discharged there in a vegetative state. Among patients with Hunt–Hess grade 4 or 5 SAH at initial presentation (n = 14), no patient had good functional status at hospital discharge, whereas seven patients had good functional status at 6 months (50 %; 95 % CI, 20–80 %). These findings were not significantly different in sensitivity analyses limited to patients with aneurysmal SAH.
Fig. 1

Trajectory of the functional status of 52 patients with non-traumatic SAH from hospital discharge to 6 months as determined by the mRS score. Circle (gray) thickness represents the proportion of patients with a specific mRS score at one point in time, and line (black) thickness represents the proportion of patients with a specific mRS score who stay or change from one time-point to the next

An exploratory multivariable analysis of possible predictors of recovery revealed several trends: hydrocephalus (OR, 4.33; 95 % CI, 0.81–23.18), clinical vasospasm (OR, 2.05; 95 % CI, 0.54–7.70), and age (OR per decade, 1.47; 95 % CI, 0.97–2.24) appeared to be associated with greater likelihood of recovery from discharge to 6 months (Table 2).
Table 2

Ordinal logistic regression analysis of clinical factors associated with functional recovery after hospital discharge for SAH

Clinical variable



OR (95 % CI)

p value

OR (95 % CI)

p value


3.60 (1.20–10.85)


4.33 (0.81–23.2)


GCS at admission

0.85 (0.73–1.00)


0.92 (0.63–1.34)


Hunt–Hess grade at admission

1.53 (0.96–2.45)


1.12 (0.39–3.45)


Clinical vasospasm

3.05 (0.85–11.00)


2.05 (0.54–7.70)


Age (decade)

1.04 (0.99–1.08)


1.47 (0.97–2.23)



2.53 (0.85–7.58)


0.48 (0.07–3.39)


Mechanical ventilation

2.52 (0.85–7.54)


0.67 (0.08–5.70)


Predictors that were significantly associated with functional recovery (i.e., any improvement in the mRS score from hospital discharge to 6 months) in univariate analysis (p value <0.10) were inserted into the final multivariate model and left in place regardless of statistical significance

GCS Glasgow Coma Scale


We have found that most patients with SAH have at least some functional recovery after hospital discharge and that more than half of those with poor functional status at discharge recover to independence at 6 months. In addition, patients with high-grade (Hunt–Hess 4 or 5) SAH who survive to hospital discharge often make considerable clinical recovery after leaving the hospital.

Patients with SAH are often critically ill throughout much of their hospitalization [1]. Therefore, the majority of these patients’ hospital stay is marked by aggressive efforts to sustain life and prevent secondary neurologic injury [1]. In such situations, physicians often defer predictions of neurologic outcome until patients are more stable. Our results may therefore be helpful to physicians who are frequently asked after a lengthy and complicated hospitalization whether a patient will ever regain functional independence.

We found trends toward independent associations between age, hydrocephalus, and clinical vasospasm and greater post-discharge recovery. These factors have been shown in prior studies to be associated with worse neurologic outcomes [5, 10, 25], so patients with these conditions would be expected to have worse functional status at discharge and thus more room for clinical recovery. However, these findings also suggest that patients with these features do recover after SAH, but at a slower pace, thus caregivers should not necessarily abandon hope for good outcomes in these patients. Nevertheless, our analysis was purely exploratory and underpowered, and no variables were significant at the multivariable level; reexamination in a larger study population is warranted to confirm these associations.

Our study benefited from its prospective, standardized, and thorough long-term followup. However, it also has several important limitations. First, our sample size was relatively small and therefore our estimates suffer from a fair degree of imprecision. This is especially true of our analyses of possible predictors of post-discharge recovery. Second, our study was conducted at a single, tertiary-care university hospital with a high volume of SAH, and therefore, our results may not be generalizable to other settings. Third, our study population was somewhat heterogeneous and was not restricted to only aneurysmal SAH. However, our main results were essentially unchanged in sensitivity analyses limited to the cases of aneurysmal SAH. Fourth, functional outcomes were only assessed by means of the mRS score, which although used frequently in SAH studies [21, 23], it was initially created for patients with ischemic stroke and is heavily weighted by patients’ ability to ambulate. We did not assess the trajectory of patients’ neuropsychological function or their ability to work, both of which are common long-term concerns in patients with SAH [3, 7]. Fifth, long-term outcomes were only evaluated at one time-point. Assessing outcomes at 3 months and at 1 year might have provided additional helpful information on the course of recovery after SAH. Sixth, we did not evaluate physiologic parameters or systematically collect data on medical events occurring between hospital discharge and 6 months. Seventh, although patients’ mRS scores at 6 months were calculated prospectively, their mRS scores at discharge and all clinical information were determined retrospectively by chart review.

In conclusion, a substantial proportion of patients with non-traumatic SAH who have the disability at discharge have significant functional recovery by 6 months. This is true even for patients with Hunt–Hess grade 4 or 5 hemorrhages. Future work is needed to determine which clinical factors predict functional recovery after SAH.

Conflicts of interest


Copyright information

© Springer Science+Business Media, LLC 2012