R. Loch Macdonald
Andreas Raabe and Bertil Romner
Kent D. Yundt, Ralph G. Dacey Jr., and Michael N. Diringer
✓ The authors reviewed clinical and financial data for all patients treated for nontraumatic subarachnoid hemorrhage (SAH) and unruptured cerebral aneurysms at their institution between June 1993 and December 1994. This study sought to identify specific areas of high resource utilization that may be amenable to reduction of expenditures without compromising quality of care. Detailed hospital financial data were correlated with clinical grade and course. Areas of high resource use were identified based on patient charges and category-specific loaded hospital cost. Patients were divided into four groups: Group 1, surgically treated unruptured aneurysms (28 patients); Group 2, acute SAH (42 patients); Group 3, SAH with vasospasm (32 patients); and Group 4, SAH with negative angiogram (10 patients). Total cost per patient (mean ± standard deviation in thousands of U.S. dollars) was highest for Group 3 (38.4 ± 21.3; vs. Group 1, 12.7 ± 8.8; Group 2, 22.6 ± 20.9; and Group 4, 25.0 ± 33.5) and correlated with hospital length of stay, Hunt and Hess grade, and Fisher grade. Areas of highest hospital cost were not always reflected in patient charges. The three areas of highest cost accounted for 48.5% of the total cost and were: 1) intensive care unit (ICU) room (Group 1, 2.5 ± 3.5; Group 2, 7.0 ± 9.2; Group 3, 11.0 ± 7.8; and Group 4, 7.9 ± 14.1); 2) arteriography (Group 1, 1.7 ± 1.2; Group 2, 2.1 ± 1.5; Group 3, 4.1 ± 2.1; and Group 4, 2.2 ± 0.7); and 3) ICU medicosurgical supplies (Group 1, 1.7 ± 0.8; Group 2, 2.0 ± 1.5; Group 3, 3.7 ± 1.7; and Group 4, 2.0 ± 3.0). It is concluded that cost containment strategies should be based on cost rather than charge and novel approaches will be required to reduce the cost of treating patients with SAH. Such approaches might include preventing vasospasm, reducing ICU stay, selective use of arteriography, and reducing the cost of supplies.
Michael N. Diringer, Jeffrey R. Kirsch, Daniel F. Hanley, and Richard J. Traystman
✓ The authors tested the hypothesis that cerebral blood flow (CBF) reactivity to CO2 was blunted following subarachnoid hemorrhage (SAH). Subarachnoid hemorrhage was produced in five cats by performing four cisterna magna injections of blood in each (SAH Group). A second group of six cats was treated with an antifibrinolytic agent (AF) in addition to four cisterna magna blood injections (SAH + AF Group). Four cats received AF and four cisterna magna injections of saline (Control Group). The presence or absence of basilar artery vasospasm was determined by comparing baseline and follow-up selective angiograms. Cerebral blood flow reactivity was determined by randomly varying the concentration of inspired CO2 to alter PaCO2 from 20 to 75 mm Hg. Regional CBF was measured with radiolabeled microspheres. Basilar artery vasospasm was seen following subarachnoid injection of blood but not of saline. Normocapnic CBF was similar in all three groups in the brain stem (mean ± standard error of the mean: SAH Group 46 ± 6, SAH + AF Group 46 ± 6, and Control Group 44 ± 9 ml/min/100 gm) and in the supratentorial compartment (SAH Group 53 ± 8, SAH + AF Group 61 ±9, and Control Group 51 ± 13 ml/min/100 gm). At intermediate levels of hypercarbia (PaCO2 50 ± 3 mm Hg), CBF increased similarly in all three groups (SAH Group 161% ± 32%, SAH + AF Group 118% ± 33%, and Control Group 174% ± 19% compared to baseline); at higher levels of PaCO2 (60 ± 3 mm Hg), CBF values were SAH Group 265% ± 50%, SAH + AF Group 205% ± 47%, and Control Group 159% ± 30% of baseline. At the highest level of PaCO2 (75 ± 6 mm Hg), supratentorial CBF did not increase as much in the SAH + AF Group as in the Control Group (179% ± 59% vs. 463% ± 58% of baseline, respectively). The authors conclude that, in this model of SAH, there is no change in normocapnic CBF; however, blood flow reactivity to hypercarbia is blunted. It is possible that this may result from a combination of narrowing of proximal large vessels and globally impaired reactivity of small vessels.
Francisco de Assis Aquino Gondim, Venkatesh Aiyagari, Angela Shackleford, and Michael N. Diringer
Mannitol is commonly used for acute insults to the central nervous system; acute renal insufficiency is one of its side effects. The cause of mannitol-induced acute renal insufficiency (MI-ARI) is unknown, although elevated osmolality has been implicated as a risk factor. The goal of this study was to determine risk factors and outcomes of MI-ARI and to determine whether osmolality is associated with MI-ARI.
The authors retrospectively reviewed the cases of 95 patients treated with mannitol to determine if MI-ARI (an increase in the creatinine level of > 0.5 mg/dl if the baseline value is < 2 mg/dl or an increase > 1 mg/dl if the baseline value is > 2 mg/dl) is linked to elevated osmolality. The 11 patients (11.6%) in whom MI-ARI developed did not exhibit significant differences in patient age, sex, or race; history of cerebrovascular disease or smoking; baseline renal function; or Glasgow Coma Scale score from those in whom MI-ARI did not occur. Cumulative fluid balance, exposure to nephrotoxic drugs, and the peak osmolality and osmotic gap before onset of renal insufficiency were also similar in the two groups. Factors predictive of the onset of MI-ARI included a higher Acute Physiology and Chronic Health Evaluation (APACHE) II score on admission and a history of diabetes, coronary artery disease, congestive heart failure, and hypertension. The presence of congestive heart failure and a high APACHE II score were the only factors independently associated with a higher likelihood of MI-ARI according to a multivariate analysis. Renal function spontaneously returned to baseline in all patients. With maintenance of normovolemia and monitoring of the osmotic gap, MI-ARI appears to be associated with chronic insults to the kidneys such as a history of diabetes or hypertension, not mannitol dose, or osmolality.
Use of osmolality to limit mannitol use and thus prevent MI-ARI may be unwarranted. Prospective studies are needed.
Rajat Dhar, Michael T. Scalfani, Allyson R. Zazulia, Tom O. Videen, Colin P. Derdeyn, and Michael N. Diringer
Critical reductions in oxygen delivery (DO2) underlie the development of delayed cerebral ischemia (DCI) after subarachnoid hemorrhage (SAH). If DO2 is not promptly restored, then irreversible injury (that is, cerebral infarction) may result. Hemodynamic therapies for DCI (that is, induced hypertension [IH] and hypervolemia) aim to improve DO2 by raising cerebral blood flow (CBF). Red blood cell (RBC) transfusion may be an alternate strategy that augments DO2 by improving arterial O2 content. The authors compared the relative ability of these 3 interventions to improve cerebral DO2, specifically their ability to restore DO2 to regions where it is impaired.
The authors compared 3 prospective physiological studies in which PET imaging was used to measure global and regional CBF and DO2 before and after the following treatments: 1) fluid bolus of 15 ml/kg normal saline (9 patients); 2) raising mean arterial pressure 25% (12 patients); and 3) transfusing 1 U of RBCs (17 patients) in 38 individuals with aneurysmal SAH at risk for DCI. Response between groups in regions with low DO2 (< 4.5 ml/100 g/min) was compared using repeated-measures ANOVA.
Groups were similar except that the fluid bolus cohort had more patients with symptoms of DCI and lower baseline CBF. Global CBF or DO2 did not rise significantly after any of the interventions, except after transfusion in patients with hemoglobin levels < 9 g/dl. All 3 treatments improved CBF and DO2 to regions with impaired baseline DO2, with a greater improvement after transfusion (23%) than hypertension (14%) or volume loading (10%); p < 0.001. Transfusion also resulted in a nonsignificantly greater (47%) reduction in the number of brain regions with low DO2 when compared with fluid bolus (7%) and hypertension (12%) (p = 0.33).
The IH, fluid bolus, and blood transfusion interventions all improve DO2 to vulnerable brain regions at risk for ischemia after SAH. Transfusion appeared to provide a physiological benefit at least comparable to IH, especially among patients with anemia, but transfusion is associated with risks. The clinical significance of these findings remains to be established in controlled clinical trials.
Michael N. Diringer, Tom O. Videen, Kent Yundt, Allyson R. Zazulia, Venkatesh Aiyagari, Ralph G. Dacey Jr., Robert L. Grubb Jr., and William J. Powers
Object. Recently, concern has been raised that hyperventilation following severe traumatic brain injury (TBI) could lead to cerebral ischemia. In acute ischemic stroke, in which the baseline metabolic rate is normal, reduction in cerebral blood flow (CBF) below a threshold of 18 to 20 ml/100 g/min is associated with energy failure. In severe TBI, however, the metabolic rate of cerebral oxygen (CMRO2) is low. The authors previously reported that moderate hyperventilation lowered global hemispheric CBF to 25 ml/100 g/min but did not alter CMRO2. In the present study they sought to determine if hyperventilation lowers CBF below the ischemic threshold of 18 to 20 ml/100 g/min in any brain region and if those reductions cause energy failure (defined as a fall in CMRO2).
Methods. Two groups of patients were studied. The moderate hyperventilation group (nine patients) underwent hyperventilation to PaCO2 of 30 ± 2 mm Hg early after TBI, regardless of intracranial pressure (ICP). The severe hyperventilation group (four patients) underwent hyperventilation to PaCO2 of 25 ± 2 mm Hg 1 to 5 days postinjury while ICP was elevated (20–30 mm Hg). The ICP, mean arterial blood pressure, and jugular venous O2 content were monitored, and cerebral perfusion pressure was maintained at 70 mm Hg or higher by using vasopressors when needed. All data are given as the mean ± standard deviation unless specified otherwise. The moderate hyperventilation group was studied 11.2 ± 1.6 hours (range 8–14 hours) postinjury, the admission Glasgow Coma Scale (GCS) score was 5.6 ± 1.8, the mean age was 27 ± 9 years, and eight of the nine patients were men. In the severe hyperventilation group, the admission GCS score was 4.3 ± 1.5, the mean age was 31 ± 6 years, and all patients were men. Positron emission tomography measurements of regional CBF, cerebral blood volume, CMRO2, and oxygen extraction fraction (OEF) were obtained before and during hyperventilation. In all 13 patients an automated search routine was used to identify 2.1-cm spherical nonoverlapping regions with CBF values below thresholds of 20, 15, and 10 ml/100 g/min during hyperventilation, and the change in CMRO2 in those regions was determined. In the regions in which CBF was less than 20 ml/100 g/min during hyperventilation, it fell from 26 ± 6.2 to 13.7 ± 1 ml/100 g/min (p < 0.0001), OEF rose from 0.31 to 0.59 (p < 0.0001), and CMRO2 was unchanged (1.12 ± 0.29 compared with 1.14 ± 0.03 ml/100 g/min; p = 0.8). In the regions in which CBF was less than 15 ml/100 g/min during hyperventilation, it fell from 23.3 ± 6.6 to 11.1 ± 1.2 ml/100 g/min (p < 0.0001), OEF rose from 0.31 to 0.63 (p < 0.0001), and CMRO2 was unchanged (0.98 ± 0.19 compared with 0.97 ± 0.23 ml/100 g/min; p = 0.92). In the regions in which CBF was less than 10 ml/100 g/min during hyperventilation, it fell from 18.2 ± 4.5 to 8.1 ± 0 ml/100 g/min (p < 0.0001), OEF rose from 0.3 to 0.71 (p < 0.0001), and CMRO2 was unchanged (0.78 ± 0.26 compared with 0.84 ± 0.32 ml/100 g/min; p = 0.64).
Conclusions. After severe TBI, brief hyperventilation produced large reductions in CBF but not energy failure, even in regions in which CBF fell below the threshold for energy failure defined in acute ischemia. Oxygen metabolism was preserved due to the low baseline metabolic rate and compensatory increases in OEF; thus, these reductions in CBF are unlikely to cause further brain injury.
James M. Milburn, Christopher J. Moran, DeWitte T. Cross III, Michael N. Diringer, Thomas K. Pilgram, and Ralph G. Dacey Jr.
Object. This study was conducted to determine if there is a change in intracranial arterial diameters after papaverine infusion for vasospasm and to determine whether the change occurs in proximal, intermediate, and distal arteries.
Methods. The authors measured arterial diameters retrospectively in all patients who received intraarterial papaverine for treatment of vasospasm between November 1992 and August 1995. Patients who received papaverine in the same session with or following angioplasty were excluded. Measurements were made in a blinded manner with the aid of a magnification loupe at 12 predetermined sites on each angiogram before and after papaverine infusion. Eighty-one treatments in 34 patients were included. Angiograms obtained at the time of presentation with subarachnoid hemorrhage (SAH) were examined in 26 of the 34 patients. Nine carotid territories visualized by repeated angiography on the day after infusion were examined to determine the duration of the papaverine effect.
Conclusions. In all treatment groups an increase was found in the average arterial diameters ranging from 2.8 to 73.9%, with a mean increase of 26.5%. Increases in diameter were observed in proximal, intermediate, and distal arteries. The timing of treatments ranged from Day 3 to Day 19 post-SAH, and there was no relationship between timing and arterial responsiveness (r = −0.06). There was a moderately good correlation between the degree of vasospasm in an artery and its responsiveness to papaverine (r = −0.54, −0.66, and −0.66, for proximal, intermediate, and distal arteries, respectively). The effect of papaverine did not persist until the following day in patients in whom repeated angiography was performed.