Journal of Critical Care xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes☆,☆☆,★ Tiffany R. Chang, MD a,⁎, Robert G. Kowalski, MS, MBBCh b, J. Ricardo Carhuapoma, MD b,c,d, Rafael J. Tamargo, MD b,d, Neeraj S. Naval, MD b,c,d a
Departments of Neurosurgery and Neurology, University of Texas Medical School, Houston, TX Anesthesia Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD d Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD b c
a r t i c l e
i n f o
Keywords: Subarachnoid hemorrhage Aneurysm Neurocritical care unit Case volume Caseload Outcome
a b s t r a c t Purpose: To compare aneurysmal subarachnoid hemorrhage (aSAH) outcomes between high- and low-volume referral centers with dedicated neurosciences critical care units (NCCUs) and shared neurosurgical, endovascular, and neurocritical care practitioners. Materials and Methods: Prospectively collected data of aSAH patients admitted to 2 institutional NCCUs were reviewed. NCCU A is a 22-bed unit staffed 24/7 with overnight in-house NCCU fellow and resident coverage. NCCU B is a 14-bed unit with home call by NCCU attending/fellow and in-house residents. Results: A total of 161 aSAH patients (27%) were admitted to NCCU B compared with 447 at NCCU A (73%). Among factors that independently impacted hospital mortality, there were no differences in baseline characteristics: mean age (A: 53.5 ± 14.1 years, B: 53.1 ± 13.6 years), poor grade Hunt and Hess (A: 28.2%, B: 26.7%), presence of multiple medical comorbidities (A: 28%, B: 31.1%), and associated cocaine use (A: 11.6%, B: 14.3%). There was no significant difference in hospital mortality (A: 17.9%, B: 18%), poor functional outcome (A: 30%, B: 25.4%), aneurysm rerupture (A: 2.8%, B: 2.4%), or delayed cerebral ischemia (A: 14.1%, B: 16.1%). Conclusions: The noninferior outcomes at the lower SAH volume center suggests that provider expertise, not patient volume, is critical to providing high-quality specialized care. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Aneurysmal subarachnoid hemorrhage (aSAH) is a relatively uncommon type of stroke that occurs in about 30 000 patients per year in the United States [1]. The management of SAH is complex and distinct from the care of the general critically ill population. A multidisciplinary team of neurointensivists, neurosurgeons, nurses, and therapists may be involved to provide high-quality care. Specialized neurosciences critical care units (NCCUs) have been shown to improve outcomes, decrease length of stay, and reduce hospital mortality in critically ill neurologic patients [2–4]. In patients with aSAH, NCCU care has been Abbreviations: aSAH, aneurysmal subarachnoid hemorrhage; ICU, intensive care unit; IMC, intermediate care unit; NCCU, neurosciences critical care unit; GCS, Glasgow Coma Scale; H/H, Hunt and Hess grade; GOS, Glasgow Outcome Scale; DCI, delayed cerebral ischemia. ☆ Acknowledgments: None. ☆☆ Sources of funding: None. ★ Conflicts of interest/disclosures: The authors have no disclosures and declare that they have no conflicts of interest. ⁎ Corresponding author at: Department of Neurosurgery, University of Texas at Houston, 6431 Fannin St., MSB 7.154, Houston, TX 77030. Tel.: +1 713 500 6128; fax: +1 713 500 0665. E-mail addresses: [email protected] (T.R. Chang), [email protected] (R.G. Kowalski), [email protected] (J.R. Carhuapoma), [email protected] (R.J. Tamargo), [email protected] (N.S. Naval).
reported to result in improved functional outcomes and a greater likelihood of discharge to home [5,6]. This is possibly attributed to more aggressive medical management, invasive monitoring, and advanced neurosurgical procedures [7]. Hospital case volume has also been suggested to impact outcomes. High-volume centers have been reported to have superior outcomes and lower mortality rates following aSAH [8–11]. Multiple components of specialized patient care besides case load may contribute to these findings, and conflicting findings of similar outcomes in high- and low-volume centers have been reported [12,13]. In the present study, we will further examine the effect of case volume on institutional aSAH outcomes by comparing 2 academic NCCUs with shared neurocritical care, neurosurgery, and endovascular providers. We hope to identify variables that may play a key role in providing high-quality specialized care and should be considered when making the decision to transfer a patient with SAH. 2. Materials and methods Patients admitted to the Johns Hopkins Medical Institutions with a diagnosis of aSAH between 2001 and 2009 were reviewed. We excluded patients with traumatic SAH, SAH secondary to an underlying lesion (eg, arteriovenous malformation, dural arteriovenous fistula, and brain
http://dx.doi.org/10.1016/j.jcrc.2015.01.007 0883-9441/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
2
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
tumor), or angiogram-negative SAH of uncertain etiology. NCCU A is a 22-bed combined intensive care unit (ICU)/intermediate care unit (IMC) staffed 24 hours/day with in-house neurocritical care fellow and resident coverage. NCCU B is a 14-bed unit (8-bed ICU/6-bed flex IMC/ICU) with a fellow on call from home with telemedicine access and in-house residents covering the general ward, emergency department consults, and the ICU/IMC. Both NCCUs are staffed overnight by attending neurointensivists who take call from home. Neurocritical care, neurosurgery, and endovascular practitioners (attendings, fellows, and residents) are shared between the 2 NCCUs. Nurses are specific to the individual hospital and do not rotate between the 2 institutions. However, both groups of nurses have received specialized training in neurosciences critical care. The 2 institutions also have similar access to diagnostic testing, radiology services, and invasive monitoring. Protocols for SAH management are similar at both institutions. Patients are admitted to the NCCU with co-management by the neurocritical care and neurosurgical teams. All patients are started on nimodipine and empiric antiepileptic drug prophylaxis with either phenytoin or levetiracetam. Angiography is obtained as soon as possible, followed by surgical clipping or endovascular coiling based on neurosurgical and neurointerventional physician review of angiographic findings. Patients are monitored with daily transcranial Doppler studies; the duration of monitoring is determined by the treating physician. Management decisions such as blood pressure control, fluid management, and vasospasm therapy are at the discretion of the individual providers. Cardiac output monitoring, invasive neuromonitoring, and continuous EEG are not routinely employed but may be used in certain cases. Baseline demographic and radiologic data were examined in patients admitted to NCCUs A and B. The impact of factors that could potentially affect outcomes such as age, admission Glasgow Coma Scale (GCS), Hunt and Hess grade (H/H), cocaine use, intraventricular hemorrhage, and medical comorbidities were analyzed. Our method of defining medical comorbidities using the Charlson index is described in detail in a prior publication [14]. Briefly, the Charlson index identifies major comorbidities including stroke, dementia, congestive heart failure, coronary and peripheral artery disease, chronic obstructive pulmonary disease, cirrhosis, peptic ulcer disease, renal failure, diabetes mellitus, human immunodeficiency virus, connective tissue disease, and cancer. This index has been validated in ischemic stroke and intracerebral hemorrhage (ICH) [15,16]. Outcome measures studied included (a) hospital mortality, (b) functional status using Glasgow Outcome Scale (GOS), and (c) complications related to aSAH including aneurysm rerupture and delayed cerebral ischemia (DCI). The detailed methodology for outcome variable review is as described in previous publications [17,18]. Poor functional outcome was defined as GOS scores 1 to 3. Glasgow Outcome Scale was measured both at discharge and at the first clinic appointment postdischarge. We determined aneurysm rerupture based on computed tomographic (CT) confirmation of new SAH, intraventricular hemorrhage, or ICH after a neurologic deterioration that prompted the CT in the first place or identified on institutional CT after interhospital transfer. Delayed cerebral ischemia was defined as (a) radiologic confirmation of cerebral infarction attributed to vasospasm or (b) symptomatic vasospasm [19]. At our institution, we defined symptomatic vasospasm as a new clinical deterioration greater than 48 hours post-SAH and greater than 24 hours after surgical clipping or endovascular coiling with either angiographic confirmation of vasospasm or clinical responsiveness (transient or sustained) to hemodynamic augmentation with transcranial Doppler corroboration. Other etiologies of neurologic deterioration, such as ventriculitis or hydrocephalus, were ruled out. 2.1. Statistical analysis An unpaired t test was used when data were normally distributed, and nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test) were used when data were not normally distributed. Dichotomous
variables were compared with outcome using the χ 2 test; the Fisher exact test result was reported where appropriate. The SPSS (Statistical Package for the Social Sciences version 18.0: SPSS Statistics, Chicago, Ill) was used to assess the potential impact of each of the admission factors on hospital mortality. The impact of various factors on outcomes was initially studied using univariate analysis; individual medical comorbidities that impacted outcomes (P b .10 on univariate analysis) were grouped together as none, single, or multiple. Finally, all factors affecting hospital mortality and outcomes were analyzed using the multiple logistic regression model. 3. Results A total of 608 patients with aSAH were included in the analysis. Four hundred forty-seven (73%) were admitted to NCCU A, for an average of 50 cases per year; 161 (27%) were admitted to NCCU B, for an average of 18 cases per year. Baseline patient characteristics which affected outcome on multivariate analysis are shown in Table 1. There were no significant differences in age based on comparison of mean age (A: 53.5 ± 14.1 years, B: 53.1 ± 13.6 years; P = .691) or advanced age N 70 years (A: 13.4%, B: 13%; P = .901). There was no significant difference in clinical severity whether assessed by comparison of proportion of poor grade H/H (A: 28.2%, B: 26.7%; P = .750) or by admission GCS b 8 (A: 21%, B: 21%; P = .981). Finally, the 2 populations did not differ significantly in terms of cocaine use at presentation (A: 11.6%, B: 14.3%; P = .400) or multiple medical comorbidities (A: 28%, B: 31.1%; P = .470). The severity of illness of the patient populations that presented to the 2 institutions with a diagnosis of aneurysmal SAH was overall similar. Annual admissions to the 2 ICUs between 2001 and 2009 are shown in Figure. 3.1. Outcome measures Overall in-hospital mortality between 2001 and 2009 was 17.9%, with a mortality rate of 41% among high-grade (H/H 4-5) patients. During this period, endovascular coiling was used in 20.9% of patients. Poor functional outcome (GOS 1-3) was seen in 28.5% of patients. The mean time to functional outcome assessment post-SAH was 37 days. Approximately 19% of hospital survivors were lost to follow-up. Outcomes from aSAH in the 2 NCCUs are shown in Table 2. No significant differences in outcome were detected between the 2 groups. Mortality rates were nearly identical between NCCU A (17.9%) and NCCU B (18%; P = 1.000). Poor functional outcome (GOS 1-3) was observed in 30% of patients in NCCU A and 25.4% of patients in NCCU B (P = .300). For survivors, this was assessed at the first clinic appointment postdischarge. While the timing of this was variable, there was no significant difference between the 2 centers when comparing mean days post-discharge for assessment. There were no differences in complications characteristically associated with aSAH including rates of aneurysm rerupture (A: 2.8%, B: 2.4%; P = .998) or DCI (A: 14.1%, B: 16.1%; P = .520) between the 2 centers. 4. Discussion We found no significant differences in aSAH outcomes between lowvolume (18 cases/y) and high-volume (50 cases/y) volume enters with Table 1 Baseline patient characteristics affecting outcomes on multivariable analysis
Age (y), mean ± SD Age N70 y Poor grade H/H (4, 5) GCS b8 Cocaine use Comorbidities: ≥2
NCCU A (n = 447)
NCCU B (n = 161)
P
53.5 ± 14.1 60 (13.4%) 126 (28.2%) 94 (21%) 52 (11.6%) 125 (28%)
53.1 ± 13.6 21 (13%) 43 (26.7%) 34 (21%) 23 (14.3%) 50 (31.1%)
.691 .901 .750 .981 .400 .470
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
80
NCCU A
NCCU B
70 60 50 40 30 20 10 0 2001
2002
2003
2004
2005
2006
2007
2008
2009
Figure. Total number of admissions per year for aSAH at NCCU A and NCCU B.
dedicated NCCUs. Similar results regarding aneurysm case load volumes have been reported in 2 large retrospective series. A study including 70 US academic medical centers reviewed outcomes from 9534 ruptured and 2623 unruptured aneurysms between 1994 and 1997 [12]. Overall case load volume was not independently associated with hospital mortality or treatment cost. A later review of aneurysm clipping outcomes in Japan in 2003 included 11974 patients [13]. Data were collected by survey, and volume was divided into 3 clipping groups: b 30, 30-50, or N 50 cases/y. No correlation was found between case volume and functional outcome at discharge for either ruptured or unruptured aneurysms. Other retrospective studies have reported improved aSAH outcomes in high-volume centers [9,11,20]. Variable definitions of “high” and “low” volume have been used, presenting a challenge in comparing different case series. Cross et al [9], who analyzed 16399 SAH admissions to US hospitals, found increased hospital mortality in the lowest volume quartile (0-9 cases) compared with the highest quartile (36-158 cases). A recent multicenter analysis in England including 23 167 patients identified a correlation between annual case volume and decreased 6-month mortality (largest center 367 cases/y) [11] Others have also reported improved stroke and SAH outcomes but higher costs and lengths of stay at large-volume centers [8,21]. Many confounding factors have to be taken into consideration which could bias these investigations, including the types of patients transferred, disease severity, and hospital mortality or withdrawal of care. When reviewed in a recent meta-analysis, high-volume centers were associated with lower hospital mortality rates following aSAH [22]. There are multiple possible contributing factors that may have led to similar aSAH outcomes between the 2 centers in our study. A major difference between this and prior studies is that the 2 sites share specialist practitioners: neurointensivists, neurosurgeons, neurointerventionalists, fellows, and residents were held constant between the 2 groups. It is therefore possible that expertise, rather than the volume of patient experience, is the major contributor to patient outcomes. Other institutional factors such as access to endovascular services and teaching hospital status have historically been reported to impact mortality [12,23]. Our study is limited by its retrospective nature and is also subject to the confounding influence of patient transfers; however, these 2 institutions share a transfer center and accept transfers based solely on bed availability and not patient characteristics or severity. Although protocols are in place, care is not standardized and varies according to the individual neurocritical care and neurosurgery providers. Patient
Table 2 Outcomes from aSAH in NCCU A compared with NCCU B
In-hospital mortality Poor functional outcome (GOS 1-3) Aneurysm rerupture DCI
NCCU A (n = 447)
NCCU B (n = 161)
P
80 (17.9%) 134 (30%) 13 (2.8%) 63 (14.1%)
29 (18%) 41 (25.4%) 4 (2.4%) 26 (16.1%)
1.000 .300 .998 .520
3
management has changed over time and will likely continue to evolve based on emerging literature. Some changes observed during this time period include hypertensive management of vasospasm over hemodilution, maintenance of euvolemia, focus on normoglycemia and normothermia, and increased availability of endovascular options. However, this would be expected to impact both centers similarly, and we believe that the comparison of the outcomes over this time frame is still valid. Another limitation is limited access to retrospective data that can compare other important end points, such as comparable costs of treatment between the institutions. Our patient population may not be representative of other centers around the country. For example, the incidence of surgical clipping is approximately 80% at our institution. Also, our DCI rates are reported to be lower than most institutions. This may, to some extent, be a reflection of our institutional definition of DCI. Finally, and perhaps most importantly, we acknowledge that our conclusions must be interpreted cautiously. We have a unique model in which our low-volume center is covered by faculty from a major academic medical center, and therefore, our findings may not be generalizable to other centers. The benefit, cost, and safety of patient transfer to a high-volume center remains controversial. However, both the American Heart Association/ American Stroke Association [1] and Neurocritical Care Society [24] guidelines for aSAH management advocate for consideration of patient transfer to a high-volume center. Ruptured aneurysms are now being treated more frequently at high-volume centers in this country, whereas the number of low-volume centers is gradually decreasing [20]. Although patient transfer to a high-volume center is certainly reasonable for many patients, given the relative paucity of subspecialized NCCU beds, restricting access of critically ill patients to lower-volume centers may not be appropriate. Such practice would be based on the flawed assumption that all low-volume centers are homogenous. Experience through volume is needed, but may be more important for the provider than the institution itself. Other factors of equal or greater significance such as specialist expertise and availability [25], endovascular capability, and academic status should all be taken into consideration, perhaps not only for triage purposes but also, by extension, even for comprehensive stroke center accreditation. 5. Conclusions We found no differences in outcome from aSAH between a low- and a high-volume center. Provider expertise, rather than just hospital case volume, may be critical for providing high-quality care and should be considered when transferring a patient with SAH. Because the structure of and relationship between the centers in this study are rather unique, our results may not be generalizable to all low-volume centers. References [1] Bederson JB, Connolly Jr ES, Batjer HH, Dacey RG, Dion JE, Diringer MN, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a statement for healthcare professionals from a special writing group of the stroke council, american heart association. Stroke 2009;40:994–1025. [2] Suarez JI, Zaidat OO, Suri MF, Feen ES, Lynch G, Hickman J, et al. Length of stay and mortality in neurocritically ill patients: impact of a specialized neurocritical care team. Crit Care Med 2004;32:2311–7. [3] Varelas PN, Eastwood D, Yun HJ, Spanaki MV, Hacein Bey L, Kessaris C, et al. Impact of a neurointensivist on outcomes in patients with head trauma treated in a neurosciences intensive care unit. J Neurosurg 2006;104:713–9. [4] Varelas PN, Schultz L, Conti M, Spanaki M, Genarrelli T, Hacein-Bey L. The impact of a neuro-intensivist on patients with stroke admitted to a neurosciences intensive care unit. Neurocrit Care 2008;9:293–9. [5] Lerch C, Yonekawa Y, Muroi C, Bjeljac M, Keller E. Specialized neurocritical care, severity grade, and outcome of patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2006;5:85–92. [6] Samuels O, Webb A, Culler S, Martin K, Barrow D. Impact of a dedicated neurocritical care team in treating patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2011;14:334–40. [7] Kurtz P, Fitts V, Sumer Z, Jalon H, Cooke J, Kvetan V, et al. How does care differ for neurological patients admitted to a neurocritical care unit versus a general icu? Neurocrit Care 2011;15:477–80. [8] Bardach NS, Zhao S, Gress DR, Lawton MT, Johnston SC. Association between subarachnoid hemorrhage outcomes and number of cases treated at California hospitals. Stroke 2002;33:1851–6.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
4
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
[9] Cross III DT, Tirschwell DL, Clark MA, Tuden D, Derdeyn CP, Moran CJ, et al. Mortality rates after subarachnoid hemorrhage: variations according to hospital case volume in 18 states. J Neurosurg 2003;99:810–7. [10] Berman MF, Solomon RA, Mayer SA, Johnston SC, Yung PP. Impact of hospital-related factors on outcome after treatment of cerebral aneurysms. Stroke 2003;34:2200–7. [11] McNeill L, English SW, Borg N, Matta BF, Menon DK. Effects of institutional caseload of subarachnoid hemorrhage on mortality: a secondary analysis of administrative data. Stroke 2013;44:647–52. [12] Johnston SC. Effect of endovascular services and hospital volume on cerebral aneurysm treatment outcomes. Stroke 2000;31:111–7. [13] Hattori N, Katayama Y, Abe T. Case volume does not correlate with outcome after cerebral aneurysm clipping: a nationwide study in Japan. Neurol Med Chir (Tokyo) 2007;47:95–100 [discussion 100–101]. [14] Naval NS, Kowalski RG, Chang TR, Caserta F, Carhuapoma JR, Tamargo RJ. The SAH score: a comprehensive communication tool. J Stroke Cerebrovasc Dis 2014;23:902–9. [15] Goldstein LB, Samsa GP, Matchar DB, Horner RD. Charlson index comorbidity adjustment for ischemic stroke outcome studies. Stroke 2004;35:1941–5 [United States]. [16] Bar B, Hemphill III JC. Charlson comorbidity index adjustment in intracerebral hemorrhage. Stroke 2011;42:2944–6 [United States]. [17] Chang TR, Kowalski RG, Caserta F, Carhuapoma JR, Tamargo RJ, Naval NS. Impact of acute cocaine use on aneurysmal subarachnoid hemorrhage. Stroke 2013;44: 1825–9.
[18] Naval NS, Chang T, Caserta F, Kowalski RG, Carhuapoma JR, Tamargo RJ. Improved aneurysmal subarachnoid hemorrhage outcomes: a comparison of 2 decades at an academic center. J Crit Care 2013;28:182–8. [19] Frontera JA, Fernandez A, Schmidt JM, Claassen J, Wartenberg KE, Badjatia N, et al. Defining vasospasm after subarachnoid hemorrhage: what is the most clinically relevant definition? Stroke 2009;40:1963–8 [United States]. [20] Leake CB, Brinjikji W, Kallmes DF, Cloft HJ. Increasing treatment of ruptured cerebral aneurysms at high-volume centers in the united states. J Neurosurg 2011;115:1179–83. [21] Tsugawa Y, Kumamaru H, Yasunaga H, Hashimoto H, Horiguchi H, Ayanian JZ. The association of hospital volume with mortality and costs of care for stroke in japan. Med Care 2013;51:782–8. [22] Boogaarts HD, van Amerongen MJ, de Vries J, Westert GP, Verbeek AL, Grotenhuis JA, et al. Caseload as a factor for outcome in aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Neurosurg 2014;120:605–11. [23] Lai PM, Lin N, Du R. Effect of teaching hospital status on outcome of aneurysm treatment. World Neurosurg 2014;82:380–5. [24] Diringer MN, Bleck TP, Claude Hemphill III J, Menon D, Shutter L, Vespa P, et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the neurocritical care society's multidisciplinary consensus conference. Neurocrit Care 2011;15:211–40. [25] Knopf L, Staff I, Gomes J, McCullough L. Impact of a neurointensivist on outcomes in critically ill stroke patients. Neurocrit Care 2012;16:63–71.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes☆,☆☆,★ Tiffany R. Chang, MD a,⁎, Robert G. Kowalski, MS, MBBCh b, J. Ricardo Carhuapoma, MD b,c,d, Rafael J. Tamargo, MD b,d, Neeraj S. Naval, MD b,c,d a
Departments of Neurosurgery and Neurology, University of Texas Medical School, Houston, TX Anesthesia Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD d Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD b c
a r t i c l e
i n f o
Keywords: Subarachnoid hemorrhage Aneurysm Neurocritical care unit Case volume Caseload Outcome
a b s t r a c t Purpose: To compare aneurysmal subarachnoid hemorrhage (aSAH) outcomes between high- and low-volume referral centers with dedicated neurosciences critical care units (NCCUs) and shared neurosurgical, endovascular, and neurocritical care practitioners. Materials and Methods: Prospectively collected data of aSAH patients admitted to 2 institutional NCCUs were reviewed. NCCU A is a 22-bed unit staffed 24/7 with overnight in-house NCCU fellow and resident coverage. NCCU B is a 14-bed unit with home call by NCCU attending/fellow and in-house residents. Results: A total of 161 aSAH patients (27%) were admitted to NCCU B compared with 447 at NCCU A (73%). Among factors that independently impacted hospital mortality, there were no differences in baseline characteristics: mean age (A: 53.5 ± 14.1 years, B: 53.1 ± 13.6 years), poor grade Hunt and Hess (A: 28.2%, B: 26.7%), presence of multiple medical comorbidities (A: 28%, B: 31.1%), and associated cocaine use (A: 11.6%, B: 14.3%). There was no significant difference in hospital mortality (A: 17.9%, B: 18%), poor functional outcome (A: 30%, B: 25.4%), aneurysm rerupture (A: 2.8%, B: 2.4%), or delayed cerebral ischemia (A: 14.1%, B: 16.1%). Conclusions: The noninferior outcomes at the lower SAH volume center suggests that provider expertise, not patient volume, is critical to providing high-quality specialized care. © 2015 Elsevier Inc. All rights reserved.
1. Introduction Aneurysmal subarachnoid hemorrhage (aSAH) is a relatively uncommon type of stroke that occurs in about 30 000 patients per year in the United States [1]. The management of SAH is complex and distinct from the care of the general critically ill population. A multidisciplinary team of neurointensivists, neurosurgeons, nurses, and therapists may be involved to provide high-quality care. Specialized neurosciences critical care units (NCCUs) have been shown to improve outcomes, decrease length of stay, and reduce hospital mortality in critically ill neurologic patients [2–4]. In patients with aSAH, NCCU care has been Abbreviations: aSAH, aneurysmal subarachnoid hemorrhage; ICU, intensive care unit; IMC, intermediate care unit; NCCU, neurosciences critical care unit; GCS, Glasgow Coma Scale; H/H, Hunt and Hess grade; GOS, Glasgow Outcome Scale; DCI, delayed cerebral ischemia. ☆ Acknowledgments: None. ☆☆ Sources of funding: None. ★ Conflicts of interest/disclosures: The authors have no disclosures and declare that they have no conflicts of interest. ⁎ Corresponding author at: Department of Neurosurgery, University of Texas at Houston, 6431 Fannin St., MSB 7.154, Houston, TX 77030. Tel.: +1 713 500 6128; fax: +1 713 500 0665. E-mail addresses: [email protected] (T.R. Chang), [email protected] (R.G. Kowalski), [email protected] (J.R. Carhuapoma), [email protected] (R.J. Tamargo), [email protected] (N.S. Naval).
reported to result in improved functional outcomes and a greater likelihood of discharge to home [5,6]. This is possibly attributed to more aggressive medical management, invasive monitoring, and advanced neurosurgical procedures [7]. Hospital case volume has also been suggested to impact outcomes. High-volume centers have been reported to have superior outcomes and lower mortality rates following aSAH [8–11]. Multiple components of specialized patient care besides case load may contribute to these findings, and conflicting findings of similar outcomes in high- and low-volume centers have been reported [12,13]. In the present study, we will further examine the effect of case volume on institutional aSAH outcomes by comparing 2 academic NCCUs with shared neurocritical care, neurosurgery, and endovascular providers. We hope to identify variables that may play a key role in providing high-quality specialized care and should be considered when making the decision to transfer a patient with SAH. 2. Materials and methods Patients admitted to the Johns Hopkins Medical Institutions with a diagnosis of aSAH between 2001 and 2009 were reviewed. We excluded patients with traumatic SAH, SAH secondary to an underlying lesion (eg, arteriovenous malformation, dural arteriovenous fistula, and brain
http://dx.doi.org/10.1016/j.jcrc.2015.01.007 0883-9441/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
2
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
tumor), or angiogram-negative SAH of uncertain etiology. NCCU A is a 22-bed combined intensive care unit (ICU)/intermediate care unit (IMC) staffed 24 hours/day with in-house neurocritical care fellow and resident coverage. NCCU B is a 14-bed unit (8-bed ICU/6-bed flex IMC/ICU) with a fellow on call from home with telemedicine access and in-house residents covering the general ward, emergency department consults, and the ICU/IMC. Both NCCUs are staffed overnight by attending neurointensivists who take call from home. Neurocritical care, neurosurgery, and endovascular practitioners (attendings, fellows, and residents) are shared between the 2 NCCUs. Nurses are specific to the individual hospital and do not rotate between the 2 institutions. However, both groups of nurses have received specialized training in neurosciences critical care. The 2 institutions also have similar access to diagnostic testing, radiology services, and invasive monitoring. Protocols for SAH management are similar at both institutions. Patients are admitted to the NCCU with co-management by the neurocritical care and neurosurgical teams. All patients are started on nimodipine and empiric antiepileptic drug prophylaxis with either phenytoin or levetiracetam. Angiography is obtained as soon as possible, followed by surgical clipping or endovascular coiling based on neurosurgical and neurointerventional physician review of angiographic findings. Patients are monitored with daily transcranial Doppler studies; the duration of monitoring is determined by the treating physician. Management decisions such as blood pressure control, fluid management, and vasospasm therapy are at the discretion of the individual providers. Cardiac output monitoring, invasive neuromonitoring, and continuous EEG are not routinely employed but may be used in certain cases. Baseline demographic and radiologic data were examined in patients admitted to NCCUs A and B. The impact of factors that could potentially affect outcomes such as age, admission Glasgow Coma Scale (GCS), Hunt and Hess grade (H/H), cocaine use, intraventricular hemorrhage, and medical comorbidities were analyzed. Our method of defining medical comorbidities using the Charlson index is described in detail in a prior publication [14]. Briefly, the Charlson index identifies major comorbidities including stroke, dementia, congestive heart failure, coronary and peripheral artery disease, chronic obstructive pulmonary disease, cirrhosis, peptic ulcer disease, renal failure, diabetes mellitus, human immunodeficiency virus, connective tissue disease, and cancer. This index has been validated in ischemic stroke and intracerebral hemorrhage (ICH) [15,16]. Outcome measures studied included (a) hospital mortality, (b) functional status using Glasgow Outcome Scale (GOS), and (c) complications related to aSAH including aneurysm rerupture and delayed cerebral ischemia (DCI). The detailed methodology for outcome variable review is as described in previous publications [17,18]. Poor functional outcome was defined as GOS scores 1 to 3. Glasgow Outcome Scale was measured both at discharge and at the first clinic appointment postdischarge. We determined aneurysm rerupture based on computed tomographic (CT) confirmation of new SAH, intraventricular hemorrhage, or ICH after a neurologic deterioration that prompted the CT in the first place or identified on institutional CT after interhospital transfer. Delayed cerebral ischemia was defined as (a) radiologic confirmation of cerebral infarction attributed to vasospasm or (b) symptomatic vasospasm [19]. At our institution, we defined symptomatic vasospasm as a new clinical deterioration greater than 48 hours post-SAH and greater than 24 hours after surgical clipping or endovascular coiling with either angiographic confirmation of vasospasm or clinical responsiveness (transient or sustained) to hemodynamic augmentation with transcranial Doppler corroboration. Other etiologies of neurologic deterioration, such as ventriculitis or hydrocephalus, were ruled out. 2.1. Statistical analysis An unpaired t test was used when data were normally distributed, and nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test) were used when data were not normally distributed. Dichotomous
variables were compared with outcome using the χ 2 test; the Fisher exact test result was reported where appropriate. The SPSS (Statistical Package for the Social Sciences version 18.0: SPSS Statistics, Chicago, Ill) was used to assess the potential impact of each of the admission factors on hospital mortality. The impact of various factors on outcomes was initially studied using univariate analysis; individual medical comorbidities that impacted outcomes (P b .10 on univariate analysis) were grouped together as none, single, or multiple. Finally, all factors affecting hospital mortality and outcomes were analyzed using the multiple logistic regression model. 3. Results A total of 608 patients with aSAH were included in the analysis. Four hundred forty-seven (73%) were admitted to NCCU A, for an average of 50 cases per year; 161 (27%) were admitted to NCCU B, for an average of 18 cases per year. Baseline patient characteristics which affected outcome on multivariate analysis are shown in Table 1. There were no significant differences in age based on comparison of mean age (A: 53.5 ± 14.1 years, B: 53.1 ± 13.6 years; P = .691) or advanced age N 70 years (A: 13.4%, B: 13%; P = .901). There was no significant difference in clinical severity whether assessed by comparison of proportion of poor grade H/H (A: 28.2%, B: 26.7%; P = .750) or by admission GCS b 8 (A: 21%, B: 21%; P = .981). Finally, the 2 populations did not differ significantly in terms of cocaine use at presentation (A: 11.6%, B: 14.3%; P = .400) or multiple medical comorbidities (A: 28%, B: 31.1%; P = .470). The severity of illness of the patient populations that presented to the 2 institutions with a diagnosis of aneurysmal SAH was overall similar. Annual admissions to the 2 ICUs between 2001 and 2009 are shown in Figure. 3.1. Outcome measures Overall in-hospital mortality between 2001 and 2009 was 17.9%, with a mortality rate of 41% among high-grade (H/H 4-5) patients. During this period, endovascular coiling was used in 20.9% of patients. Poor functional outcome (GOS 1-3) was seen in 28.5% of patients. The mean time to functional outcome assessment post-SAH was 37 days. Approximately 19% of hospital survivors were lost to follow-up. Outcomes from aSAH in the 2 NCCUs are shown in Table 2. No significant differences in outcome were detected between the 2 groups. Mortality rates were nearly identical between NCCU A (17.9%) and NCCU B (18%; P = 1.000). Poor functional outcome (GOS 1-3) was observed in 30% of patients in NCCU A and 25.4% of patients in NCCU B (P = .300). For survivors, this was assessed at the first clinic appointment postdischarge. While the timing of this was variable, there was no significant difference between the 2 centers when comparing mean days post-discharge for assessment. There were no differences in complications characteristically associated with aSAH including rates of aneurysm rerupture (A: 2.8%, B: 2.4%; P = .998) or DCI (A: 14.1%, B: 16.1%; P = .520) between the 2 centers. 4. Discussion We found no significant differences in aSAH outcomes between lowvolume (18 cases/y) and high-volume (50 cases/y) volume enters with Table 1 Baseline patient characteristics affecting outcomes on multivariable analysis
Age (y), mean ± SD Age N70 y Poor grade H/H (4, 5) GCS b8 Cocaine use Comorbidities: ≥2
NCCU A (n = 447)
NCCU B (n = 161)
P
53.5 ± 14.1 60 (13.4%) 126 (28.2%) 94 (21%) 52 (11.6%) 125 (28%)
53.1 ± 13.6 21 (13%) 43 (26.7%) 34 (21%) 23 (14.3%) 50 (31.1%)
.691 .901 .750 .981 .400 .470
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
80
NCCU A
NCCU B
70 60 50 40 30 20 10 0 2001
2002
2003
2004
2005
2006
2007
2008
2009
Figure. Total number of admissions per year for aSAH at NCCU A and NCCU B.
dedicated NCCUs. Similar results regarding aneurysm case load volumes have been reported in 2 large retrospective series. A study including 70 US academic medical centers reviewed outcomes from 9534 ruptured and 2623 unruptured aneurysms between 1994 and 1997 [12]. Overall case load volume was not independently associated with hospital mortality or treatment cost. A later review of aneurysm clipping outcomes in Japan in 2003 included 11974 patients [13]. Data were collected by survey, and volume was divided into 3 clipping groups: b 30, 30-50, or N 50 cases/y. No correlation was found between case volume and functional outcome at discharge for either ruptured or unruptured aneurysms. Other retrospective studies have reported improved aSAH outcomes in high-volume centers [9,11,20]. Variable definitions of “high” and “low” volume have been used, presenting a challenge in comparing different case series. Cross et al [9], who analyzed 16399 SAH admissions to US hospitals, found increased hospital mortality in the lowest volume quartile (0-9 cases) compared with the highest quartile (36-158 cases). A recent multicenter analysis in England including 23 167 patients identified a correlation between annual case volume and decreased 6-month mortality (largest center 367 cases/y) [11] Others have also reported improved stroke and SAH outcomes but higher costs and lengths of stay at large-volume centers [8,21]. Many confounding factors have to be taken into consideration which could bias these investigations, including the types of patients transferred, disease severity, and hospital mortality or withdrawal of care. When reviewed in a recent meta-analysis, high-volume centers were associated with lower hospital mortality rates following aSAH [22]. There are multiple possible contributing factors that may have led to similar aSAH outcomes between the 2 centers in our study. A major difference between this and prior studies is that the 2 sites share specialist practitioners: neurointensivists, neurosurgeons, neurointerventionalists, fellows, and residents were held constant between the 2 groups. It is therefore possible that expertise, rather than the volume of patient experience, is the major contributor to patient outcomes. Other institutional factors such as access to endovascular services and teaching hospital status have historically been reported to impact mortality [12,23]. Our study is limited by its retrospective nature and is also subject to the confounding influence of patient transfers; however, these 2 institutions share a transfer center and accept transfers based solely on bed availability and not patient characteristics or severity. Although protocols are in place, care is not standardized and varies according to the individual neurocritical care and neurosurgery providers. Patient
Table 2 Outcomes from aSAH in NCCU A compared with NCCU B
In-hospital mortality Poor functional outcome (GOS 1-3) Aneurysm rerupture DCI
NCCU A (n = 447)
NCCU B (n = 161)
P
80 (17.9%) 134 (30%) 13 (2.8%) 63 (14.1%)
29 (18%) 41 (25.4%) 4 (2.4%) 26 (16.1%)
1.000 .300 .998 .520
3
management has changed over time and will likely continue to evolve based on emerging literature. Some changes observed during this time period include hypertensive management of vasospasm over hemodilution, maintenance of euvolemia, focus on normoglycemia and normothermia, and increased availability of endovascular options. However, this would be expected to impact both centers similarly, and we believe that the comparison of the outcomes over this time frame is still valid. Another limitation is limited access to retrospective data that can compare other important end points, such as comparable costs of treatment between the institutions. Our patient population may not be representative of other centers around the country. For example, the incidence of surgical clipping is approximately 80% at our institution. Also, our DCI rates are reported to be lower than most institutions. This may, to some extent, be a reflection of our institutional definition of DCI. Finally, and perhaps most importantly, we acknowledge that our conclusions must be interpreted cautiously. We have a unique model in which our low-volume center is covered by faculty from a major academic medical center, and therefore, our findings may not be generalizable to other centers. The benefit, cost, and safety of patient transfer to a high-volume center remains controversial. However, both the American Heart Association/ American Stroke Association [1] and Neurocritical Care Society [24] guidelines for aSAH management advocate for consideration of patient transfer to a high-volume center. Ruptured aneurysms are now being treated more frequently at high-volume centers in this country, whereas the number of low-volume centers is gradually decreasing [20]. Although patient transfer to a high-volume center is certainly reasonable for many patients, given the relative paucity of subspecialized NCCU beds, restricting access of critically ill patients to lower-volume centers may not be appropriate. Such practice would be based on the flawed assumption that all low-volume centers are homogenous. Experience through volume is needed, but may be more important for the provider than the institution itself. Other factors of equal or greater significance such as specialist expertise and availability [25], endovascular capability, and academic status should all be taken into consideration, perhaps not only for triage purposes but also, by extension, even for comprehensive stroke center accreditation. 5. Conclusions We found no differences in outcome from aSAH between a low- and a high-volume center. Provider expertise, rather than just hospital case volume, may be critical for providing high-quality care and should be considered when transferring a patient with SAH. Because the structure of and relationship between the centers in this study are rather unique, our results may not be generalizable to all low-volume centers. References [1] Bederson JB, Connolly Jr ES, Batjer HH, Dacey RG, Dion JE, Diringer MN, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a statement for healthcare professionals from a special writing group of the stroke council, american heart association. Stroke 2009;40:994–1025. [2] Suarez JI, Zaidat OO, Suri MF, Feen ES, Lynch G, Hickman J, et al. Length of stay and mortality in neurocritically ill patients: impact of a specialized neurocritical care team. Crit Care Med 2004;32:2311–7. [3] Varelas PN, Eastwood D, Yun HJ, Spanaki MV, Hacein Bey L, Kessaris C, et al. Impact of a neurointensivist on outcomes in patients with head trauma treated in a neurosciences intensive care unit. J Neurosurg 2006;104:713–9. [4] Varelas PN, Schultz L, Conti M, Spanaki M, Genarrelli T, Hacein-Bey L. The impact of a neuro-intensivist on patients with stroke admitted to a neurosciences intensive care unit. Neurocrit Care 2008;9:293–9. [5] Lerch C, Yonekawa Y, Muroi C, Bjeljac M, Keller E. Specialized neurocritical care, severity grade, and outcome of patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2006;5:85–92. [6] Samuels O, Webb A, Culler S, Martin K, Barrow D. Impact of a dedicated neurocritical care team in treating patients with aneurysmal subarachnoid hemorrhage. Neurocrit Care 2011;14:334–40. [7] Kurtz P, Fitts V, Sumer Z, Jalon H, Cooke J, Kvetan V, et al. How does care differ for neurological patients admitted to a neurocritical care unit versus a general icu? Neurocrit Care 2011;15:477–80. [8] Bardach NS, Zhao S, Gress DR, Lawton MT, Johnston SC. Association between subarachnoid hemorrhage outcomes and number of cases treated at California hospitals. Stroke 2002;33:1851–6.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
4
T.R. Chang et al. / Journal of Critical Care xxx (2015) xxx–xxx
[9] Cross III DT, Tirschwell DL, Clark MA, Tuden D, Derdeyn CP, Moran CJ, et al. Mortality rates after subarachnoid hemorrhage: variations according to hospital case volume in 18 states. J Neurosurg 2003;99:810–7. [10] Berman MF, Solomon RA, Mayer SA, Johnston SC, Yung PP. Impact of hospital-related factors on outcome after treatment of cerebral aneurysms. Stroke 2003;34:2200–7. [11] McNeill L, English SW, Borg N, Matta BF, Menon DK. Effects of institutional caseload of subarachnoid hemorrhage on mortality: a secondary analysis of administrative data. Stroke 2013;44:647–52. [12] Johnston SC. Effect of endovascular services and hospital volume on cerebral aneurysm treatment outcomes. Stroke 2000;31:111–7. [13] Hattori N, Katayama Y, Abe T. Case volume does not correlate with outcome after cerebral aneurysm clipping: a nationwide study in Japan. Neurol Med Chir (Tokyo) 2007;47:95–100 [discussion 100–101]. [14] Naval NS, Kowalski RG, Chang TR, Caserta F, Carhuapoma JR, Tamargo RJ. The SAH score: a comprehensive communication tool. J Stroke Cerebrovasc Dis 2014;23:902–9. [15] Goldstein LB, Samsa GP, Matchar DB, Horner RD. Charlson index comorbidity adjustment for ischemic stroke outcome studies. Stroke 2004;35:1941–5 [United States]. [16] Bar B, Hemphill III JC. Charlson comorbidity index adjustment in intracerebral hemorrhage. Stroke 2011;42:2944–6 [United States]. [17] Chang TR, Kowalski RG, Caserta F, Carhuapoma JR, Tamargo RJ, Naval NS. Impact of acute cocaine use on aneurysmal subarachnoid hemorrhage. Stroke 2013;44: 1825–9.
[18] Naval NS, Chang T, Caserta F, Kowalski RG, Carhuapoma JR, Tamargo RJ. Improved aneurysmal subarachnoid hemorrhage outcomes: a comparison of 2 decades at an academic center. J Crit Care 2013;28:182–8. [19] Frontera JA, Fernandez A, Schmidt JM, Claassen J, Wartenberg KE, Badjatia N, et al. Defining vasospasm after subarachnoid hemorrhage: what is the most clinically relevant definition? Stroke 2009;40:1963–8 [United States]. [20] Leake CB, Brinjikji W, Kallmes DF, Cloft HJ. Increasing treatment of ruptured cerebral aneurysms at high-volume centers in the united states. J Neurosurg 2011;115:1179–83. [21] Tsugawa Y, Kumamaru H, Yasunaga H, Hashimoto H, Horiguchi H, Ayanian JZ. The association of hospital volume with mortality and costs of care for stroke in japan. Med Care 2013;51:782–8. [22] Boogaarts HD, van Amerongen MJ, de Vries J, Westert GP, Verbeek AL, Grotenhuis JA, et al. Caseload as a factor for outcome in aneurysmal subarachnoid hemorrhage: a systematic review and meta-analysis. J Neurosurg 2014;120:605–11. [23] Lai PM, Lin N, Du R. Effect of teaching hospital status on outcome of aneurysm treatment. World Neurosurg 2014;82:380–5. [24] Diringer MN, Bleck TP, Claude Hemphill III J, Menon D, Shutter L, Vespa P, et al. Critical care management of patients following aneurysmal subarachnoid hemorrhage: recommendations from the neurocritical care society's multidisciplinary consensus conference. Neurocrit Care 2011;15:211–40. [25] Knopf L, Staff I, Gomes J, McCullough L. Impact of a neurointensivist on outcomes in critically ill stroke patients. Neurocrit Care 2012;16:63–71.
Please cite this article as: Chang TR, et al, Impact of case volume on aneurysmal subarachnoid hemorrhage outcomes, J Crit Care (2015), http:// dx.doi.org/10.1016/j.jcrc.2015.01.007
