Background and Purpose The sudden death rate from aneurysmal subarachnoid hemorrhage (SAH) is 10%. Since 1989, 26 SAH patients who were witnessed to collapse into coma with respiratory arrest and required cardiopulmonary resuscitation (CPR) at the scene survived to reach the hospital and be diagnosed. Although reports on hospital management of grade V SAH suggest improved outcome, no report has previously addressed the issue of respiratory arrest after acute SAH. We analyze our experience with this unique subgroup of aneurysmal SAH patients.
Methods This is a retrospective analysis of 26 consecutive SAH patients who collapsed at the scene and required CPR for respiratory arrest and survived to reach the hospital and be diagnosed. Statistical analysis was performed using the t test and Mann-Whitney rank-sum test.
Results All patients were grade V on arrival at the emergency department. Twenty-one patients received mouth-to-mouth resuscitation only, and 5 received chest compressions as well. The mean duration of bystander CPR was 12 to 15 minutes. CT scan showed diffuse, thick SAH in all patients, an associated subdural hemorrhage in 2, and an intraparenchymal hemorrhage in 4. After CT scan, an intracranial pressure (ICP) monitor was placed in 24, and 2 were taken to emergency surgery for subdural and intracerebral hemorrhage. ICP was elevated in 24 patients (mean, 54 mm Hg), and a ventriculostomy was placed in all 24. ICP was unresponsive in 12, and all suffered brain death. ICP lessened to <25 mm Hg in 12, and all underwent angiography. All 12 had an aneurysm and underwent emergency surgical clipping. Time to surgery from SAH was ≤11 hours in all 12 patients. All were managed with calcium channel blockers and hyperdynamic therapy in addition to aggressive control of ICP. The outcome at 12 months in the 14 surgical cases was normal in 3 patients (21%), good in 2 (14%), vegetative in 1 (7%), and death in 8 (57%).
Conclusions Aneurysmal SAH patients that present with respiratory arrest present as grade V patients with elevated ICP. Bystander CPR coupled with early retrieval, diagnosis, and therapy can lead to 20% functional survival in what used to be sudden death from aneurysmal SAH.
The incidence of sudden death from aneurysmal SAH is difficult to determine precisely. It appears that the mortality in the first 24 hours in North America is approximately 10%.1 2 With an acute aneurysmal rupture and sudden elevation of ICP, some patients immediately lapse into coma and go into respiratory arrest. Some of these apoplectic events are witnessed. The exact percentage of our population in the United States that is capable of performing CPR is not known, but with the tremendous educational effort by the AHA in our schools, workplaces, and the media, we suspect that the majority of the American population knows CPR or at least mouth-to-mouth resuscitation.3 Additionally, emergency medical response programs such as the 911 emergency telephone number can lead to rapid emergency response and retrieval to the hospital by highly trained paramedics. Our institution is an American College of Surgeons–certified level 1 trauma center with the ability to do around-the-clock emergency CT scans and angiograms with no delay. Our institution also operates an ambulance and paramedic program and is certified in the instruction of prehospital trauma life support, advanced cardiac life support, and advanced trauma life support. Because of this, we have had the unique experience of receiving a significant number of acute aneurysmal SAH patients who were resuscitated with bystander CPR at the scene of the rupture. Although there are reports on the management of poor-grade aneurysm patients, no report has addressed this unique population of patients; thus, we have analyzed our experience with SAH patients who suffered a witnessed respiratory arrest.
Patient Population and Results
Between 1989 and 1994, 26 consecutive SAH patients who acutely collapsed into coma with subsequent respiratory arrest were resuscitated with bystander CPR. The location of the apoplectic event was in the workplace in 11 instances, in the home in 6, in a car in 3, in our hospital in 2, in a store in 2, and on the street in 2. The mean duration from time of notification of 911 services until paramedic arrival was 12 minutes, with the longest response being 17 minutes. Verification of who administered the CPR was available in 21 of 26 cases. Mouth-to-mouth resuscitation was given in all instances, and in 6 instances chest compressions were also performed. Of the 21 verified instances, 6 bystanders performing CPR had formal CPR training, and the remainder remembered watching instructional films on CPR sponsored by the AHA on multiple occasions in school. In all 26 cases, the paramedics were told that the patient was believed to not be breathing when CPR was initiated. Emergency medical paramedics intubated all 26 patients at the scene and transported them by ambulance to our emergency department. There was no instance of cardiac arrhythmia requiring defibrillation or cardioversion. On arrival to the emergency department, no patient was in cardiac arrest, but 5 were profoundly hypotensive and required fluid resuscitation and inotropic support. Of the profoundly hypotensive patients, only case 1 (Table 3⇓) survived. All other patients were either normotensive or hypertensive. Using the World Federation of Neurological Surgeons SAH grading scale, all 26 patients were grade V. Chest x-ray revealed neurogenic pulmonary edema in 4 of 5 hypotensive cases and 3 of 21 nonhypotensive cases. Emergency CT scan was obtained in all 26 cases, documenting a diffuse, thick SAH in all the patients. Two patients were taken directly to surgery for large subdural hematoma with associated intraparenchymal hematoma; both had a catastrophic rupture from a middle cerebral aneurysm (Table 1⇓). The other 24 cases had a fiberoptic intraparenchymal pressure monitor placed that documented an elevated ICP in all cases. Hyperventilation and mannitol were administered and a ventriculostomy was placed in all 24 cases. In 12 cases (mean ICP, 60), the ICP responded minimally and transiently to the therapy with no clinical improvement (Table 2⇓). No additional effort at diagnosis was made, and all 12 patients suffered brain death within 48 hours. Only four autopsies were granted, and all documented a ruptured aneurysm.
In 12 cases (mean ICP, 43), the ICP responded with some evidence for clinical improvement (Table 3⇓). These patients were taken for emergency angiography. On demonstration of the aneurysm, all were taken to surgery for emergency early clipping. Every effort was made to remove as much clot from the subarachnoid space. In 9 cases, this included irrigation with 10 mg tissue plasminogen activator, using the methods reported by Findlay et al.4 Seven of 9 cases demonstrated marked clearing of the subarachnoid blood on follow-up CT scan. All of the patients had marked elevations in their transcranial Doppler flow velocities, and all survivors required continuation of their ventriculostomy cerebrospinal fluid drainage for at least 1 week after surgery. Aggressive hyperdynamic therapy was used postoperatively in addition to calcium channel blockers and endovascular therapy as necessary. Of the 5 good-outcome (measured with Glasgow Outcome Scale) survivors, 2 returned to full-time employment.
With use of both a t test and Mann-Whitney rank-sum test to compare ICP, MAP, and cerebral perfusion pressure between the groups, Tables 2 and 3⇑⇑ reveal a significant difference for ICP between the two groups (P=.005) but no significant difference for MAP (P=.824) or cerebral perfusion pressure (P=.15)
All of the patients that suffered an acute respiratory arrest from aneurysmal SAH were grade V patients with diffuse, thick SAH and marked elevations in their ICP. These 26 patients were 70% of all the grade V patients seen at our institution during this same period. The classic work by Thompson and Malina5 documented that dynamic axial brain stem distortion from acute ICP pressure differentials between the supratentorial compartment and infratentorial compartment can lead to respiratory arrest and that this is the most likely explanation for the respiratory arrest in these patients. We feel that these patients would have become “sudden deaths” if it were not for the bystander CPR and early medical response. It is well documented that rescue breathing with mouth-to-mouth technique is beneficial to apneic patients without cardiac arrest.6 Bailes et al7 documented intracranial hypertension in the vast majority of poor-grade aneurysm patients. They also documented that when early reduction of the intracranial hypertension is achieved with clinical improvement, then early aneurysm clipping and aggressive treatment of the ensuing vasospasm can lead to improved functional survival. In the patients who did not respond to ventriculostomy, all went on to brain death within 48 hours, which is identical to our experience. Bailes et al7 documented a 22% good/normal outcome for grade V aneurysmal SAH. LeRoux et al8 documented a 24.1% favorable outcome in grade V aneurysmal SAH. Our 19% good/normal outcome is similar to these reports. All of these reports also had a 75% to 92% rate of intraventricular hemorrhage in addition to the SAH in the grade V patients. No mention of CPR was made in the reports by Bailes et al and LeRoux et al. Vegetative/poor survival is not a desirable end point of this approach, and we did withdraw support from cases 9 and 11 in Table 3⇑, both of whom were felt to be medically futile cases destined for a very poor outcome. Vegetative survival following aggressive management of grade V SAH ranged from 4% in this report to 10% in the report of Bailes et al. During the period of this experience, part of the intraoperative therapy was the use of intracisternal tissue plasminogen activator, but it appears that this has no effect on outcome, and we have since abandoned its use.9
In conclusion, it appears that there is a strong association between grade V SAH, acutely elevated ICP, and sudden respiratory arrest. We believe that unwitnessed respiratory arrest after aneurysmal SAH will lead to sudden death in most instances. Patients with witnessed respiratory arrest after SAH can be successfully resuscitated with CPR and treated with approximately 20% functional survival.
Selected Abbreviations and Acronyms
|AHA||=||American Heart Association|
|MAP||=||mean arterial pressure|
Reviews of this manuscript were directed by Hermes Kontos, MD.
- Received February 9, 1996.
- Revision received May 20, 1996.
- Accepted June 13, 1996.
- Copyright © 1996 by American Heart Association
Phillips L, Whisnant J, O'Fallon W, Sundt T. The unchanging pattern of subarachnoid hemorrhage in a community. Neurology. 1980;30:1034-1040.
Sahs A, Perret G, Locksley H, Nishioka H, eds. Intracranial Aneurysms and Subarachnoid Hemorrhage: A Cooperative Study. Philadelphia, Pa: Lippincott; 1969.
Cobb L, Eliastam M, Kerber R, Melker R, Moss A, Newell L, Paraskos J, Weaver W, Weil M, Weisfeldt M. Report of the American Heart Association Task Force on the future of cardiopulmonary resuscitation. Circulation. 1992;85:2346-2355.