- Email: [email protected]
Primary Pontine Hemorrhage in the Acute Stage: Clinical Features and a Proposed New Simple Scoring System
Primary Pontine Hemorrhage in the Acute Stage: Clinical Features and a Proposed New Simple Scoring System Toshinari Meguro, MD, Ken Kuwahara, MD, Yusuke Tomita, MD, Yu Okuma, MD, Tomoyuki Tanabe, MD, Kenichiro Muraoka, MD, Kinya Terada, MD, Nobuyuki Hirotsune, MD, and Shigeki Nishino, MD
Background: It is important to evaluate the likelihood of fatality in patients with acute primary pontine hemorrhage (PPH) in emergency departments. We aimed to evaluate the clinical symptoms and computed tomography findings of PPH to develop a simple grading scale for predicting the mortality of PPH. Methods: Records of 101 consecutive patients admitted to our hospital with acute PPH between June 1, 2006, and January 31, 2014, were retrospectively reviewed. Independent predictors of 30-day mortality were identified by univariate and multivariate logistic regression analyses. A simple and easy clinical score (PPH score) was developed from independent factors to predict mortality in acute PPH. The PPH score was compared with the established intracerebral hemorrhage (ICH) score, which served as the reference scoring system. Results: Overall mortality rate 30 days after onset was 58.4% (59 of 101). Factors independently associated with 30-day mortality were Glasgow Coma Scale (GCS) score of 6 or less (P 5 .0051), absence of pupillary light reflex (P 5 .0003), and blood glucose of 180 mg/dL or greater (P 5 .0312). The PPH score was the sum of independent factors, which were assigned 1 point each. The area under the receiver operating characteristic curve for predicting 30-day mortality was .90 (95% confidence interval [CI], .84-.95) for PPH score and .86 (95% CI, .78-.93) for ICH score. Conclusions: GCS score of 6 or less, absence of pupillary light reflex, and plasma glucose of 10 mmol/L or greater are independent mortality predictors of PPH. The PPH score is a simple and reliable clinical grading scale for predicting 30-day mortality. Key Words: Primary pontine hemorrhage— outcome—prognosis—mortality. Ó 2015 by National Stroke Association
Primary pontine hemorrhage (PPH) is known to have one of the worst prognoses among the anatomically defined subtypes of intracerebral hemorrhage (ICH). The
From the Department of Neurological Surgery, Hiroshima City Hospital, Hiroshima, Japan. Received September 12, 2014; revision received November 23, 2014; accepted December 3, 2014. The authors declare that they have no conflict of interest. Address correspondence to Toshinari Meguro, MD, Department of Neurological Surgery, Hiroshima City Hospital, 7-33 Motomachi, Naka-ku, Hiroshima 730-8518, Japan. E-mail: tmeguron@hotmail. com. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.12.006
860
prognostic implications of various clinical and computed tomography (CT) parameters have been characterized in patients with PPH,1-8 but appropriate treatment options for severe PPH, which are determined by these distinct parameters have not yet been developed. Thus, it is extremely important that physicians in the emergency department evaluate the prognostic state of acute stage PPH to judge the necessity of limited neurointensive care. Several grading scales for the evaluation of ICH have been designed to date,9-14 but there are no specific grading scales for PPH. Moreover, the ICH score has proven to be reliable in predicting 30-day mortality throughout various affected regions of the brain.9,12 The ICH score is based on analysis of 152 ICH patients,
Journal of Stroke and Cerebrovascular Diseases, Vol. 24, No. 4 (April), 2015: pp 860-865
PPH AND A NEW SCORING SYSTEM
but only 15 patients with pontine hemorrhage were included in the development of this score.9 In this study, we analyzed the clinical features of PPH from 101 consecutive cases in our hospital. Based on our findings, we propose a new and simple grading system for predicting mortality from PPH based on acute stage characteristics.
Materials and Methods We retrospectively studied a total of 101 consecutive patients with acute PPH who had been referred to our hospital between June 2006 and January 2014. All cases of arteriovenous malformation, cryptic angioma, and other intratumor hemorrhage were excluded. They represented 11.5% of the 881 patients with ICH seen at our clinic during this period. Their mean age was 63.3 years (range, 37-91 years), and 67 patients (67.3% of the total) were men. Among the 101 patients, there were 64 patients with the history of hypertension and 21 patients without the history of hypertension, and 16 patients were unidentified. As for the history of diabetes, there were 23 patients with diabetic history and 59 patients without diabetic history, and 19 patients were unidentified. All patients were admitted to our stroke unit within 24 hours of symptom onset and diagnosed by CT within 30 minutes of admission. Endotracheal intubation and artificial respirator management were not positively performed, except for the case when the emergency steps and the family at the time of a visit to the hospital wish. It was performed for 26 patients, and 20 patients died within 30 days among them. None of the patients in the study underwent surgical evacuation of hematoma or continuous ventricular drainage to treat ventricular dilatation. A follow-up examination was conducted 30 days after onset. Outcomes were assessed according to the modified Rankin Scale (mRS). Patients were divided into 2 simple groups: a survival group (mRS, 0-5) and a death group (mRS, 6). Then, we statistically evaluated the difference between the 2 groups, using parameters made up of clinical manifestations and CT findings. We assessed age, sex, the initial level of consciousness (based on the Glasgow Coma Scale [GCS]), the absence of pupillary light reflex, systolic blood pressure, heart rate, white blood cell counts, blood glucose level, extension of hematoma, intraventricular hematoma, ventricular dilatation, maximum size of hematoma, and volume of hematoma. We assessed the maximum diameter of hematoma on the horizontal section of CT and the vertical extension of the hematoma with respect to the midbrain and/or thalamus or subthalamus. The total volume of the hemorrhage was estimated using the formula for an ellipsoid, that is, 4/3 p abc, where a, b, and c represent the respective radii in 3 dimensions.15 A new outcome risk stratification scale, that we now name the PPH score, was developed through analysis of
861
the variables associated with 30-day mortality in all PPH patients whom we studied. Cut points of the variables were chosen to produce a simple and intuitive model and to incorporate values similar to those used in prior reports.9,10 The PPH score was compared with the established ICH score,9 which we used as the reference scoring system. We analyzed the variables by univariate comparisons; the chi-square test or the Fisher exact test for categorical variables and Mann–Whitney U test for continuous variables. Parameters proven to be significant on univariate comparison were analyzed again with multivariate logistic regression analysis to identify those variables that independently associated with 30-day mortality. To evaluate the prediction accuracy, we used the receiver operator characteristic curves of the ICH score and the PPH score of 30-day mortality, in addition to the corresponding area under the curve. All P values were 2 sided and P values of .05 or less were considered statistically significant. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics.16
Results Analysis of Clinical and CT Findings Total mortality rate at 30 days after onset was 58.4% (59 of 101). One patient died from renal failure on day 23, one patient died from pneumonia on day 23, and the other 56 patients died from primary brain damage within 14 days. The clinical features and CT findings in all 101 patients were analyzed. Their mean age was 63 6 14 years (range, 37-91 years), 67 patients (66.3%) were men. Mean GCS score on admission was 7 6 4 (range, 3-15). Pupillary light reflex was absent in 49 patients (48.5%). Mean systolic blood pressure on hospital arrival was 203 6 45 mm Hg (range, 55-364 mm Hg) and mean heart rate was 98 6 24 bpm (range, 52-180 bpm). Highest body temperature during the first 3 hours after admission was obtained in 93 patients, with a mean of 38.6 6 .3 C (range, 35.1-41.7 C). White blood cell count was obtained in 100 patients, with a mean of 10.1 6 3.5 3 109/L (range, 3.7-19.7 3 109/L). Blood glucose level was obtained in 99 patients with a mean of 181.7 6 82.8 mg/dL (range, 93.6-567 mg/dL). Mean maximum diameter of hematoma was 29 6 9 mm (range, 5-46 mm). Fifty-eight patients (57.4%) had intraventricular hemorrhage and 15 (14.9%) had ventricular dilatation. Hematoma was within the pons in 38 patients (37.6%). Extension of the hematoma into midbrain was found in 39 patients (38.6%), and 24 patients (23.8%) had extension of the hematoma into thalamus.
T. MEGURO ET AL.
862
Table 1. Univariate analysis of characteristics of 101 cases of primary pontine hemorrhage
Variable
N
30-Day mortality, n (%)
Age, y , 60 47 $ 60 54 Sex Man 68 Woman 33 GCS #6 68 $7 33 Light reflex (1) 52 (-) 49 SBP, mm Hg ,180 30 $180 69 HR, bpm ,100 51 $100 47 Highest body temperature, C ,39 51 $39 42 WBC, /mL ,10.000 55 $10,000 45 BG, mg/dL ,180 61 $180 38 Extension of hematoma Pons 38 Midbrain 39 Mesencephalon 24 IVH (-) 43 (1) 58 Maximum diameter, mm ,20 19 $20 82 Volume, mL , 20 83 $20 18
P
32 (68) 27 (50)
.0727
44 (65) 15 (45)
.0857
55 (81) 4 (12)
,.0001
15 (29) 44 (90)
,.0001
13 (43) 43 (62)
.0792
23 (45) 33 (70)
.0148
10 (20) 24 (57)
,.0001
29 (53) 29 (64)
.3090
27 (44) 30 (79)
.0008
14 (37) 22 (56) 23 (96)
,.0001
14 (33) 45 (78)
,.0001
1 (5) 58 (71)
,.0001
42 (51) 17 (94)
.0004
Abbreviations: BG, blood glucose; GCS, Glasgow Coma Scale; HR, heart rate; IVH, intraventricular hemorrhage; SBP, systolic blood pressure; WBC, white blood cell.
The comparisons of these parameters with univariate analysis are shown in Table 1. GCS score of 6 or less (P , .0001), the absence of pupillary light reflex (P , .0001), heart rate of 100 bpm or more (P 5 .0148), body temperature of 39 C or more (P , .0001), blood glucose level of 180 mg/dL or greater (P 5 .0008), extension of hematoma (P , .0001), the presence of intraventricular hematoma (P , .0001), maximum diameter of hematoma of 20 mm or more (P , .0001), and volume of
Table 2. Multivariate analysis of significant independent predictors of 30-day mortality after primary pontine hemorrhage Variable
Odds ratio (95% CI)
P
GCS, #6 Light reflex, (-) BG, $180 mg/dL
7.44 (1.83-30.30) 15.2 (3.53-65.40) 4.49 (1.15-17.60)
.0051 .0003 .0312
Abbreviations: BG, blood glucose; CI, confidence interval; GCS, Glasgow Coma Scale.
hematoma of 20 mL or greater (P 5 .0004) were all associated with 30-day mortality. Age, sex, systolic blood pressure, and white blood cell counts were not associated with mortality. Multivariate analysis identified GCS score of 6 or less (P 5.0051), the absence of pupillary light reflex (P 5.0003), and blood glucose of 180 mg/dL or more (P 5 .0312) as significant independent parameters (Table 2).
Development of Grading System for PPH A new and simple outcome scale of PPH was developed from the 3 characteristics that were determined to be independent predictors of 30-day mortality by multivariate regression analysis in our study. GCS score of 6 or less, the absence of pupillary light reflex, and blood glucose of 180 mg/dL or greater were each assigned 1 point; the PPH score is derived by summing the points for each predictive factor. The lowest score possible is 0 and the highest score possible is 3. The PPH score was compared with the established ICH score used as the reference scoring system. The ICH score is composed of 5 factors: GCS score (2 points for scores of 3 or 4, 1 point for scores of 5-12, and 0 point for scores of 13-15), age (1 point for patients $80 years and 0 points for patients ,80 years), ICH volume (1 point for hemorrhages $30 mL and 0 points for those ,30 mL), ventricular blood (1 point), and infratentorial bleeding (1 point).9 We applied the PPH and ICH scores to our patients, the PPH scores and the ICH scores, along with the 30-day outcome, are shown in Figure 1. The PPH score indicated 30-day mortality rates of 7.7%, 33.3%, 78.9%, and 100% for patients with 0, 1, 2, and 3 points, respectively. Using a cutoff value of $2 points for mortality, the score had a sensitivity of 87.7% (95% confidence interval [CI], 76.3%-94.9%), a specificity of 81% (95% CI, 65.9%-91.4%), a positive predictive value of 86.2% (95% CI, 74.6%-93.9%), and a negative predictive value of 82.9% (95% CI, 67.9%-92.8%). The ICH score indicated 30-day mortality rates of 10.5%, 31.8%, 73.1%, 90.9%, and 91.7% for patients with 1, 2, 3, 4, and 5 points, respectively. Using a cutoff value of $3 points for mortality (as in the original publication),9 the score had a sensitivity of 81.6% (95% CI, 68%-91.2%), a specificity of 76.2% (95% CI, 60.5%-87.9%), a positive predictive value of
PPH AND A NEW SCORING SYSTEM
863
Figure 1. Thirty-day outcome corresponding to PPH and ICH scores. Abbreviations: ICH, intracerebral hemorrhage; mRS, modified Rankin Scale; PPH, primary pontine hemorrhage.
80% (95% CI, 66.3%-90%), and a negative predictive value of 78% (95% CI, 62.4%-89.4%). Figure 2 shows the receiver operating characteristic curves of the PPH score and the ICH score in predicting 30-day mortality. No differences were found in area under the curve between the PPH score and the ICH score (P 5 .211).
Discussion PPH accounts for about 10% of ICH. Even in recent studies, the prognosis was found to be highly fatal with the overall mortality rate being 40%-60%1-5,7,8,17 because treatment is usually conservative and surgical therapy is not recommended.18 As a result, there are no definitive treatment options for severe PPH. However, there are
Figure 2. Receiver operating characteristic curves of PPH and ICH scores for predicting 30-day mortality. Abbreviations: ICH, intracerebral hemorrhage; PPH, primary pontine hemorrhage.
many surviving cases of PPH with or without severe neurologic deficits. An accurate prognostic assessment on admission is critical for establishing a reasonable therapeutic approach.2 The results of the present study show by multivariate logistic analysis that GCS score of 6 or less, the absence of pupillary light reflex, and blood glucose level of 180 mg/dL or greater are independent mortality predictors of acute PPF. In agreement with our findings, recent studies predict fatal outcome in patients with coma on admission, pupillary abnormalities, tachycardia, large hematomas, intraventricular extension, hydrocephalus, and additional factors.1-5,7,8 Among these predictors, all past reports indicate that coma (ie, consciousness disturbance) on admission is the most reliable predictive factor for poor outcome of PPH. Disturbances of consciousness on admission could be explained by rapid and severe brain stem destruction because of the effects of hemorrhage on the reticular activating system in the brain stem and/or thalamus. Thus, the level of consciousness is a major predictor of PPH outcome. Various eye signs (eg, pinpoint pupils, medial longitudinal fasciculus syndrome, one and a half syndrome, ocular bobbing, and skew deviation) have been observed in acute PPH.19 Recent studies indicate that pupillary abnormalities (anisocoria, pinpoint pupils, and mydriasis),8 dilated pupils,4 and the absence of oculocephalic response1,2 are significantly related to mortality. Many of the eye signs associated with PPH indicate damage to the brain stem by hematoma, but it is difficult for physicians who do not specialize in neurology to diagnose eye symptoms correctly in the emergency room. We therefore focused on the pupillary light reflex. The absence of pupillary light reflex in PPH suggests dysfunction of the brain stem and is simple even for residents and nurses to judge. Several studies have shown that hyperglycemia is associated with increased early mortality and poor functional outcome in diabetic and nondiabetic patients
T. MEGURO ET AL.
864 20-22
with ICH. Although, to our knowledge, past reports of PPH did not explore blood glucose levels. Hyperglycemia in the setting of acute neurologic injury is attributed to a catecholamine surge and the generalized stress response, among other factors.23 The mechanisms by which hyperglycemia increases mortality and impedes neurologic recovery are not clear. Animal studies suggest that there are deleterious effects of hyperglycemia on the brain after cerebral ischemia, which may be attributed to its secondary effects of acidosis, release of excitatory amino acids, altered inflammatory response including release of inflammatory cytokines and tumor necrosis factor a, impairment of the capillary integrity, increased blood-brain barrier permeability due to greater free radical formation and bradykinin release, and exacerbation of edema formation and hemorrhagic transformation of the damaged brain tissue.24-28 An additional clinical cause of increased glucose level was a history of or the presence of risk factors for diabetes mellitus.21 Hyperglycemia at admission and during the clinical course may also increase the susceptibility to various inhospital complications, such as cardiac failure, respiratory failure, and nosocomial infections, all of which are contributors to poor outcome.29 There are no specific grading scales for predicting mortality or functional outcome of PPH. Of the previously developed scores to predict favorable outcome versus unfavorable outcome or death after ICH,9-14 the ICH score has been widely validated.9,12 In addition, Del Brutto et al17 showed that ICH score is accurate for predicting mortality in patients with PPH. Our result also shows that the ICH score is highly accurate in predicting 30-day mortality of acute PPH patients. Hemphill et al9 have noted that, to be easily applicable and effective, a clinical grading scale must be simple enough to use without significant special training, statistical knowledge, or an extensive time commitment. It must also be reliable in patient stratification and should be composed of elements that are associated with outcome and that would likely be assessed as part of general, routine clinical care. The ICH score meets these criteria in predicting mortality from PPH. However, the PPH score, which we present here, is an even simpler grading scale composed of only 3 factors: GCS, papillary light reflex, and blood glucose level. It does not even require measurement of the volume of hematoma, as does the ICH score. Therefore, when we see PPH patients in the emergency department, we can easily and rapidly assess the 30-day mortality using the PPH score with identical accuracy to the ICH score. There are several limitations to this study. First, our study was retrospective. Second, it was performed at a single institute and the patients were all Japanese of Asian ethnicity. Third, we did not consider early do-not resuscitate orders for severe PPH. It is possible that do-not resuscitate order might induce overestimation of the 30-day
mortality of PPH. We need a prospective and more comprehensive study that accounts for ethical issues to overcome these limitations. In conclusion, GCS score of 6 or less, the absence of pupillary light reflex, and plasma glucose of 180 mg/dL or greater are independent mortality predictors of PPH. The PPH score, developed using these independent predictors, is a simple and excellent clinical grading scale for predicting the 30-day mortality.
References 1. Wijdicks EU, St Louis E. Clinical profiles predictive of outcome in pontine hemorrhage. Neurology 1997; 49:1342-1346. 2. Dziewas R, Kremer M, Ludemann P, et al. The prognostic impact of clinical and CT parameters in patients with pontine hemorrhage. Cerebrovasc Dis 2003;16:224-229. 3. Wessels T, Moller-Hartmann W, Noth J, et al. CT findings and clinical features as markers for patient outcome in primary pontine hemorrhage. AJNR Am J Neuroradiol 2004;25:257-260. 4. Jang JH, Song YG, Kim YZ. Predictors of 30-day mortality and 90-day functional recovery after primary pontine hemorrhage. J Korean Med Sci 2011;26:100-107. 5. Balci K, Asil T, Kerimoglu M, et al. Clinical and neuroradiological predictors of mortality in patients with primary pontine hemorrhage. Clin Neurol Neurosurg 2005;108:36-39. 6. Rabinstein AA, Tisch SH, McClelland RL, et al. Cause is the main predictor of outcome in patients with pontine hemorrhage. Cerebrovasc Dis 2004;17:66-71. 7. Jung DS, Jeon BC, Park YS, et al. The predictors of survival and functional outcome in patients with pontine hemorrhage. J Korean Neurosurg Soc 2007;41:82-87. 8. Murata Y, Yamaguchi S, Kajikawa H, et al. Relationship between the clinical manifestations, computed tomographic findings and the outcome in 80 patients with primary pontine hemorrhage. J Neurol Sci 1999;167:107-111. 9. Hemphill JC 3rd, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke 2001;32:891-897. 10. Weimar C, Benemann J, Diener HC. Development and validation of the Essen intracerebral hemorrhage score. J Neurol Neurosurg Psychiatry 2006;77:601-605. 11. Shaya M, Dubey A, Berk C, et al. Factors influencing outcome in intracerebral hematoma: a simple, reliable, and accurate method to grade intracerebral hemorrhage. Surg Neurol 2005;63:343-348. discussion 348. 12. Cheung RT, Zou LY. Use of the original, modified, or new intracerebral hemorrhage score to predict mortality and morbidity after intracerebral hemorrhage. Stroke 2003; 34:1717-1722. 13. Ruiz-Sandoval JL, Chiquete E, Romero-Vargas S, et al. Grading scale for prediction of outcome in primary intracerebral hemorrhages. Stroke 2007;38:1641-1644. 14. Beslow LA, Ichord RN, Gindville MC, et al. Pediatric intracerebral hemorrhage score: a simple grading scale for intracerebral hemorrhage in children. Stroke 2014;45:66-70. 15. Broderick JP, Brott TG, Duldner JE, et al. Volume of intracerebral hemorrhage. a powerful and easy-to-use predictor of 30-day mortality. Stroke 1993;24:987-993. 16. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 2013;48:452-458.
PPH AND A NEW SCORING SYSTEM 17. Del Brutto OH, Campos X. Validation of intracerebral hemorrhage scores for patients with pontine hemorrhage. Neurology 2004;62:515-516. 18. Morgenstern LB, Hemphill JC 3rd, Anderson C, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2010;41:2108-2129. 19. Nakajima K. Clinicopathological study of pontine hemorrhage. Stroke 1983;14:485-493. 20. Stead LG, Jain A, Bellolio MF, et al. Emergency department hyperglycemia as a predictor of early mortality and worse functional outcome after intracerebral hemorrhage. Neurocrit Care 2010;13:67-74. 21. Kimura K, Iguchi Y, Inoue T, et al. Hyperglycemia independently increases the risk of early death in acute spontaneous intracerebral hemorrhage. J Neurol Sci 2007;255:90-94. 22. Tetri S, Juvela S, Saloheimo P, et al. Hypertension and diabetes as predictors of early death after spontaneous intracerebral hemorrhage. J Neurosurg 2009;110:411-417. 23. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke 2001;32:2426-2432.
865 24. Allen CL, Bayraktutan U. Antioxidants attenuate hyperglycaemia-mediated brain endothelial cell dysfunction and blood-brain barrier hyperpermeability. Diabetes Obes Metab 2009;11:480-490. 25. Kamada H, Yu F, Nito C, et al. Influence of hyperglycemia on oxidative stress and matrix metalloproteinase-9 activation after focal cerebral ischemia/reperfusion in rats: relation to blood-brain barrier dysfunction. Stroke 2007; 38:1044-1049. 26. Liu J, Gao BB, Clermont AC, et al. Hyperglycemiainduced cerebral hematoma expansion is mediated by plasma kallikrein. Nat Med 2011;17:206-210. 27. Esposito KI, Nappo F, Marfella R, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 2002;106:2067-2072. 28. Martini SR, Kent TA. Hyperglycemia in acute ischemic stroke: a vascular perspective. J Cereb Blood Flow Metab 2007;27:435-451. 29. Kruyt ND, Biessels GJ, DeVries JH, et al. Hyperglycemia in aneurysmal subarachnoid hemorrhage: a potentially modifiable risk factor for poor outcome. J Cereb Blood Flow Metab 2010;30:1577-1587.
Background: It is important to evaluate the likelihood of fatality in patients with acute primary pontine hemorrhage (PPH) in emergency departments. We aimed to evaluate the clinical symptoms and computed tomography findings of PPH to develop a simple grading scale for predicting the mortality of PPH. Methods: Records of 101 consecutive patients admitted to our hospital with acute PPH between June 1, 2006, and January 31, 2014, were retrospectively reviewed. Independent predictors of 30-day mortality were identified by univariate and multivariate logistic regression analyses. A simple and easy clinical score (PPH score) was developed from independent factors to predict mortality in acute PPH. The PPH score was compared with the established intracerebral hemorrhage (ICH) score, which served as the reference scoring system. Results: Overall mortality rate 30 days after onset was 58.4% (59 of 101). Factors independently associated with 30-day mortality were Glasgow Coma Scale (GCS) score of 6 or less (P 5 .0051), absence of pupillary light reflex (P 5 .0003), and blood glucose of 180 mg/dL or greater (P 5 .0312). The PPH score was the sum of independent factors, which were assigned 1 point each. The area under the receiver operating characteristic curve for predicting 30-day mortality was .90 (95% confidence interval [CI], .84-.95) for PPH score and .86 (95% CI, .78-.93) for ICH score. Conclusions: GCS score of 6 or less, absence of pupillary light reflex, and plasma glucose of 10 mmol/L or greater are independent mortality predictors of PPH. The PPH score is a simple and reliable clinical grading scale for predicting 30-day mortality. Key Words: Primary pontine hemorrhage— outcome—prognosis—mortality. Ó 2015 by National Stroke Association
Primary pontine hemorrhage (PPH) is known to have one of the worst prognoses among the anatomically defined subtypes of intracerebral hemorrhage (ICH). The
From the Department of Neurological Surgery, Hiroshima City Hospital, Hiroshima, Japan. Received September 12, 2014; revision received November 23, 2014; accepted December 3, 2014. The authors declare that they have no conflict of interest. Address correspondence to Toshinari Meguro, MD, Department of Neurological Surgery, Hiroshima City Hospital, 7-33 Motomachi, Naka-ku, Hiroshima 730-8518, Japan. E-mail: tmeguron@hotmail. com. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2014.12.006
860
prognostic implications of various clinical and computed tomography (CT) parameters have been characterized in patients with PPH,1-8 but appropriate treatment options for severe PPH, which are determined by these distinct parameters have not yet been developed. Thus, it is extremely important that physicians in the emergency department evaluate the prognostic state of acute stage PPH to judge the necessity of limited neurointensive care. Several grading scales for the evaluation of ICH have been designed to date,9-14 but there are no specific grading scales for PPH. Moreover, the ICH score has proven to be reliable in predicting 30-day mortality throughout various affected regions of the brain.9,12 The ICH score is based on analysis of 152 ICH patients,
Journal of Stroke and Cerebrovascular Diseases, Vol. 24, No. 4 (April), 2015: pp 860-865
PPH AND A NEW SCORING SYSTEM
but only 15 patients with pontine hemorrhage were included in the development of this score.9 In this study, we analyzed the clinical features of PPH from 101 consecutive cases in our hospital. Based on our findings, we propose a new and simple grading system for predicting mortality from PPH based on acute stage characteristics.
Materials and Methods We retrospectively studied a total of 101 consecutive patients with acute PPH who had been referred to our hospital between June 2006 and January 2014. All cases of arteriovenous malformation, cryptic angioma, and other intratumor hemorrhage were excluded. They represented 11.5% of the 881 patients with ICH seen at our clinic during this period. Their mean age was 63.3 years (range, 37-91 years), and 67 patients (67.3% of the total) were men. Among the 101 patients, there were 64 patients with the history of hypertension and 21 patients without the history of hypertension, and 16 patients were unidentified. As for the history of diabetes, there were 23 patients with diabetic history and 59 patients without diabetic history, and 19 patients were unidentified. All patients were admitted to our stroke unit within 24 hours of symptom onset and diagnosed by CT within 30 minutes of admission. Endotracheal intubation and artificial respirator management were not positively performed, except for the case when the emergency steps and the family at the time of a visit to the hospital wish. It was performed for 26 patients, and 20 patients died within 30 days among them. None of the patients in the study underwent surgical evacuation of hematoma or continuous ventricular drainage to treat ventricular dilatation. A follow-up examination was conducted 30 days after onset. Outcomes were assessed according to the modified Rankin Scale (mRS). Patients were divided into 2 simple groups: a survival group (mRS, 0-5) and a death group (mRS, 6). Then, we statistically evaluated the difference between the 2 groups, using parameters made up of clinical manifestations and CT findings. We assessed age, sex, the initial level of consciousness (based on the Glasgow Coma Scale [GCS]), the absence of pupillary light reflex, systolic blood pressure, heart rate, white blood cell counts, blood glucose level, extension of hematoma, intraventricular hematoma, ventricular dilatation, maximum size of hematoma, and volume of hematoma. We assessed the maximum diameter of hematoma on the horizontal section of CT and the vertical extension of the hematoma with respect to the midbrain and/or thalamus or subthalamus. The total volume of the hemorrhage was estimated using the formula for an ellipsoid, that is, 4/3 p abc, where a, b, and c represent the respective radii in 3 dimensions.15 A new outcome risk stratification scale, that we now name the PPH score, was developed through analysis of
861
the variables associated with 30-day mortality in all PPH patients whom we studied. Cut points of the variables were chosen to produce a simple and intuitive model and to incorporate values similar to those used in prior reports.9,10 The PPH score was compared with the established ICH score,9 which we used as the reference scoring system. We analyzed the variables by univariate comparisons; the chi-square test or the Fisher exact test for categorical variables and Mann–Whitney U test for continuous variables. Parameters proven to be significant on univariate comparison were analyzed again with multivariate logistic regression analysis to identify those variables that independently associated with 30-day mortality. To evaluate the prediction accuracy, we used the receiver operator characteristic curves of the ICH score and the PPH score of 30-day mortality, in addition to the corresponding area under the curve. All P values were 2 sided and P values of .05 or less were considered statistically significant. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics.16
Results Analysis of Clinical and CT Findings Total mortality rate at 30 days after onset was 58.4% (59 of 101). One patient died from renal failure on day 23, one patient died from pneumonia on day 23, and the other 56 patients died from primary brain damage within 14 days. The clinical features and CT findings in all 101 patients were analyzed. Their mean age was 63 6 14 years (range, 37-91 years), 67 patients (66.3%) were men. Mean GCS score on admission was 7 6 4 (range, 3-15). Pupillary light reflex was absent in 49 patients (48.5%). Mean systolic blood pressure on hospital arrival was 203 6 45 mm Hg (range, 55-364 mm Hg) and mean heart rate was 98 6 24 bpm (range, 52-180 bpm). Highest body temperature during the first 3 hours after admission was obtained in 93 patients, with a mean of 38.6 6 .3 C (range, 35.1-41.7 C). White blood cell count was obtained in 100 patients, with a mean of 10.1 6 3.5 3 109/L (range, 3.7-19.7 3 109/L). Blood glucose level was obtained in 99 patients with a mean of 181.7 6 82.8 mg/dL (range, 93.6-567 mg/dL). Mean maximum diameter of hematoma was 29 6 9 mm (range, 5-46 mm). Fifty-eight patients (57.4%) had intraventricular hemorrhage and 15 (14.9%) had ventricular dilatation. Hematoma was within the pons in 38 patients (37.6%). Extension of the hematoma into midbrain was found in 39 patients (38.6%), and 24 patients (23.8%) had extension of the hematoma into thalamus.
T. MEGURO ET AL.
862
Table 1. Univariate analysis of characteristics of 101 cases of primary pontine hemorrhage
Variable
N
30-Day mortality, n (%)
Age, y , 60 47 $ 60 54 Sex Man 68 Woman 33 GCS #6 68 $7 33 Light reflex (1) 52 (-) 49 SBP, mm Hg ,180 30 $180 69 HR, bpm ,100 51 $100 47 Highest body temperature, C ,39 51 $39 42 WBC, /mL ,10.000 55 $10,000 45 BG, mg/dL ,180 61 $180 38 Extension of hematoma Pons 38 Midbrain 39 Mesencephalon 24 IVH (-) 43 (1) 58 Maximum diameter, mm ,20 19 $20 82 Volume, mL , 20 83 $20 18
P
32 (68) 27 (50)
.0727
44 (65) 15 (45)
.0857
55 (81) 4 (12)
,.0001
15 (29) 44 (90)
,.0001
13 (43) 43 (62)
.0792
23 (45) 33 (70)
.0148
10 (20) 24 (57)
,.0001
29 (53) 29 (64)
.3090
27 (44) 30 (79)
.0008
14 (37) 22 (56) 23 (96)
,.0001
14 (33) 45 (78)
,.0001
1 (5) 58 (71)
,.0001
42 (51) 17 (94)
.0004
Abbreviations: BG, blood glucose; GCS, Glasgow Coma Scale; HR, heart rate; IVH, intraventricular hemorrhage; SBP, systolic blood pressure; WBC, white blood cell.
The comparisons of these parameters with univariate analysis are shown in Table 1. GCS score of 6 or less (P , .0001), the absence of pupillary light reflex (P , .0001), heart rate of 100 bpm or more (P 5 .0148), body temperature of 39 C or more (P , .0001), blood glucose level of 180 mg/dL or greater (P 5 .0008), extension of hematoma (P , .0001), the presence of intraventricular hematoma (P , .0001), maximum diameter of hematoma of 20 mm or more (P , .0001), and volume of
Table 2. Multivariate analysis of significant independent predictors of 30-day mortality after primary pontine hemorrhage Variable
Odds ratio (95% CI)
P
GCS, #6 Light reflex, (-) BG, $180 mg/dL
7.44 (1.83-30.30) 15.2 (3.53-65.40) 4.49 (1.15-17.60)
.0051 .0003 .0312
Abbreviations: BG, blood glucose; CI, confidence interval; GCS, Glasgow Coma Scale.
hematoma of 20 mL or greater (P 5 .0004) were all associated with 30-day mortality. Age, sex, systolic blood pressure, and white blood cell counts were not associated with mortality. Multivariate analysis identified GCS score of 6 or less (P 5.0051), the absence of pupillary light reflex (P 5.0003), and blood glucose of 180 mg/dL or more (P 5 .0312) as significant independent parameters (Table 2).
Development of Grading System for PPH A new and simple outcome scale of PPH was developed from the 3 characteristics that were determined to be independent predictors of 30-day mortality by multivariate regression analysis in our study. GCS score of 6 or less, the absence of pupillary light reflex, and blood glucose of 180 mg/dL or greater were each assigned 1 point; the PPH score is derived by summing the points for each predictive factor. The lowest score possible is 0 and the highest score possible is 3. The PPH score was compared with the established ICH score used as the reference scoring system. The ICH score is composed of 5 factors: GCS score (2 points for scores of 3 or 4, 1 point for scores of 5-12, and 0 point for scores of 13-15), age (1 point for patients $80 years and 0 points for patients ,80 years), ICH volume (1 point for hemorrhages $30 mL and 0 points for those ,30 mL), ventricular blood (1 point), and infratentorial bleeding (1 point).9 We applied the PPH and ICH scores to our patients, the PPH scores and the ICH scores, along with the 30-day outcome, are shown in Figure 1. The PPH score indicated 30-day mortality rates of 7.7%, 33.3%, 78.9%, and 100% for patients with 0, 1, 2, and 3 points, respectively. Using a cutoff value of $2 points for mortality, the score had a sensitivity of 87.7% (95% confidence interval [CI], 76.3%-94.9%), a specificity of 81% (95% CI, 65.9%-91.4%), a positive predictive value of 86.2% (95% CI, 74.6%-93.9%), and a negative predictive value of 82.9% (95% CI, 67.9%-92.8%). The ICH score indicated 30-day mortality rates of 10.5%, 31.8%, 73.1%, 90.9%, and 91.7% for patients with 1, 2, 3, 4, and 5 points, respectively. Using a cutoff value of $3 points for mortality (as in the original publication),9 the score had a sensitivity of 81.6% (95% CI, 68%-91.2%), a specificity of 76.2% (95% CI, 60.5%-87.9%), a positive predictive value of
PPH AND A NEW SCORING SYSTEM
863
Figure 1. Thirty-day outcome corresponding to PPH and ICH scores. Abbreviations: ICH, intracerebral hemorrhage; mRS, modified Rankin Scale; PPH, primary pontine hemorrhage.
80% (95% CI, 66.3%-90%), and a negative predictive value of 78% (95% CI, 62.4%-89.4%). Figure 2 shows the receiver operating characteristic curves of the PPH score and the ICH score in predicting 30-day mortality. No differences were found in area under the curve between the PPH score and the ICH score (P 5 .211).
Discussion PPH accounts for about 10% of ICH. Even in recent studies, the prognosis was found to be highly fatal with the overall mortality rate being 40%-60%1-5,7,8,17 because treatment is usually conservative and surgical therapy is not recommended.18 As a result, there are no definitive treatment options for severe PPH. However, there are
Figure 2. Receiver operating characteristic curves of PPH and ICH scores for predicting 30-day mortality. Abbreviations: ICH, intracerebral hemorrhage; PPH, primary pontine hemorrhage.
many surviving cases of PPH with or without severe neurologic deficits. An accurate prognostic assessment on admission is critical for establishing a reasonable therapeutic approach.2 The results of the present study show by multivariate logistic analysis that GCS score of 6 or less, the absence of pupillary light reflex, and blood glucose level of 180 mg/dL or greater are independent mortality predictors of acute PPF. In agreement with our findings, recent studies predict fatal outcome in patients with coma on admission, pupillary abnormalities, tachycardia, large hematomas, intraventricular extension, hydrocephalus, and additional factors.1-5,7,8 Among these predictors, all past reports indicate that coma (ie, consciousness disturbance) on admission is the most reliable predictive factor for poor outcome of PPH. Disturbances of consciousness on admission could be explained by rapid and severe brain stem destruction because of the effects of hemorrhage on the reticular activating system in the brain stem and/or thalamus. Thus, the level of consciousness is a major predictor of PPH outcome. Various eye signs (eg, pinpoint pupils, medial longitudinal fasciculus syndrome, one and a half syndrome, ocular bobbing, and skew deviation) have been observed in acute PPH.19 Recent studies indicate that pupillary abnormalities (anisocoria, pinpoint pupils, and mydriasis),8 dilated pupils,4 and the absence of oculocephalic response1,2 are significantly related to mortality. Many of the eye signs associated with PPH indicate damage to the brain stem by hematoma, but it is difficult for physicians who do not specialize in neurology to diagnose eye symptoms correctly in the emergency room. We therefore focused on the pupillary light reflex. The absence of pupillary light reflex in PPH suggests dysfunction of the brain stem and is simple even for residents and nurses to judge. Several studies have shown that hyperglycemia is associated with increased early mortality and poor functional outcome in diabetic and nondiabetic patients
T. MEGURO ET AL.
864 20-22
with ICH. Although, to our knowledge, past reports of PPH did not explore blood glucose levels. Hyperglycemia in the setting of acute neurologic injury is attributed to a catecholamine surge and the generalized stress response, among other factors.23 The mechanisms by which hyperglycemia increases mortality and impedes neurologic recovery are not clear. Animal studies suggest that there are deleterious effects of hyperglycemia on the brain after cerebral ischemia, which may be attributed to its secondary effects of acidosis, release of excitatory amino acids, altered inflammatory response including release of inflammatory cytokines and tumor necrosis factor a, impairment of the capillary integrity, increased blood-brain barrier permeability due to greater free radical formation and bradykinin release, and exacerbation of edema formation and hemorrhagic transformation of the damaged brain tissue.24-28 An additional clinical cause of increased glucose level was a history of or the presence of risk factors for diabetes mellitus.21 Hyperglycemia at admission and during the clinical course may also increase the susceptibility to various inhospital complications, such as cardiac failure, respiratory failure, and nosocomial infections, all of which are contributors to poor outcome.29 There are no specific grading scales for predicting mortality or functional outcome of PPH. Of the previously developed scores to predict favorable outcome versus unfavorable outcome or death after ICH,9-14 the ICH score has been widely validated.9,12 In addition, Del Brutto et al17 showed that ICH score is accurate for predicting mortality in patients with PPH. Our result also shows that the ICH score is highly accurate in predicting 30-day mortality of acute PPH patients. Hemphill et al9 have noted that, to be easily applicable and effective, a clinical grading scale must be simple enough to use without significant special training, statistical knowledge, or an extensive time commitment. It must also be reliable in patient stratification and should be composed of elements that are associated with outcome and that would likely be assessed as part of general, routine clinical care. The ICH score meets these criteria in predicting mortality from PPH. However, the PPH score, which we present here, is an even simpler grading scale composed of only 3 factors: GCS, papillary light reflex, and blood glucose level. It does not even require measurement of the volume of hematoma, as does the ICH score. Therefore, when we see PPH patients in the emergency department, we can easily and rapidly assess the 30-day mortality using the PPH score with identical accuracy to the ICH score. There are several limitations to this study. First, our study was retrospective. Second, it was performed at a single institute and the patients were all Japanese of Asian ethnicity. Third, we did not consider early do-not resuscitate orders for severe PPH. It is possible that do-not resuscitate order might induce overestimation of the 30-day
mortality of PPH. We need a prospective and more comprehensive study that accounts for ethical issues to overcome these limitations. In conclusion, GCS score of 6 or less, the absence of pupillary light reflex, and plasma glucose of 180 mg/dL or greater are independent mortality predictors of PPH. The PPH score, developed using these independent predictors, is a simple and excellent clinical grading scale for predicting the 30-day mortality.
References 1. Wijdicks EU, St Louis E. Clinical profiles predictive of outcome in pontine hemorrhage. Neurology 1997; 49:1342-1346. 2. Dziewas R, Kremer M, Ludemann P, et al. The prognostic impact of clinical and CT parameters in patients with pontine hemorrhage. Cerebrovasc Dis 2003;16:224-229. 3. Wessels T, Moller-Hartmann W, Noth J, et al. CT findings and clinical features as markers for patient outcome in primary pontine hemorrhage. AJNR Am J Neuroradiol 2004;25:257-260. 4. Jang JH, Song YG, Kim YZ. Predictors of 30-day mortality and 90-day functional recovery after primary pontine hemorrhage. J Korean Med Sci 2011;26:100-107. 5. Balci K, Asil T, Kerimoglu M, et al. Clinical and neuroradiological predictors of mortality in patients with primary pontine hemorrhage. Clin Neurol Neurosurg 2005;108:36-39. 6. Rabinstein AA, Tisch SH, McClelland RL, et al. Cause is the main predictor of outcome in patients with pontine hemorrhage. Cerebrovasc Dis 2004;17:66-71. 7. Jung DS, Jeon BC, Park YS, et al. The predictors of survival and functional outcome in patients with pontine hemorrhage. J Korean Neurosurg Soc 2007;41:82-87. 8. Murata Y, Yamaguchi S, Kajikawa H, et al. Relationship between the clinical manifestations, computed tomographic findings and the outcome in 80 patients with primary pontine hemorrhage. J Neurol Sci 1999;167:107-111. 9. Hemphill JC 3rd, Bonovich DC, Besmertis L, et al. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke 2001;32:891-897. 10. Weimar C, Benemann J, Diener HC. Development and validation of the Essen intracerebral hemorrhage score. J Neurol Neurosurg Psychiatry 2006;77:601-605. 11. Shaya M, Dubey A, Berk C, et al. Factors influencing outcome in intracerebral hematoma: a simple, reliable, and accurate method to grade intracerebral hemorrhage. Surg Neurol 2005;63:343-348. discussion 348. 12. Cheung RT, Zou LY. Use of the original, modified, or new intracerebral hemorrhage score to predict mortality and morbidity after intracerebral hemorrhage. Stroke 2003; 34:1717-1722. 13. Ruiz-Sandoval JL, Chiquete E, Romero-Vargas S, et al. Grading scale for prediction of outcome in primary intracerebral hemorrhages. Stroke 2007;38:1641-1644. 14. Beslow LA, Ichord RN, Gindville MC, et al. Pediatric intracerebral hemorrhage score: a simple grading scale for intracerebral hemorrhage in children. Stroke 2014;45:66-70. 15. Broderick JP, Brott TG, Duldner JE, et al. Volume of intracerebral hemorrhage. a powerful and easy-to-use predictor of 30-day mortality. Stroke 1993;24:987-993. 16. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 2013;48:452-458.
PPH AND A NEW SCORING SYSTEM 17. Del Brutto OH, Campos X. Validation of intracerebral hemorrhage scores for patients with pontine hemorrhage. Neurology 2004;62:515-516. 18. Morgenstern LB, Hemphill JC 3rd, Anderson C, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2010;41:2108-2129. 19. Nakajima K. Clinicopathological study of pontine hemorrhage. Stroke 1983;14:485-493. 20. Stead LG, Jain A, Bellolio MF, et al. Emergency department hyperglycemia as a predictor of early mortality and worse functional outcome after intracerebral hemorrhage. Neurocrit Care 2010;13:67-74. 21. Kimura K, Iguchi Y, Inoue T, et al. Hyperglycemia independently increases the risk of early death in acute spontaneous intracerebral hemorrhage. J Neurol Sci 2007;255:90-94. 22. Tetri S, Juvela S, Saloheimo P, et al. Hypertension and diabetes as predictors of early death after spontaneous intracerebral hemorrhage. J Neurosurg 2009;110:411-417. 23. Capes SE, Hunt D, Malmberg K, et al. Stress hyperglycemia and prognosis of stroke in nondiabetic and diabetic patients: a systematic overview. Stroke 2001;32:2426-2432.
865 24. Allen CL, Bayraktutan U. Antioxidants attenuate hyperglycaemia-mediated brain endothelial cell dysfunction and blood-brain barrier hyperpermeability. Diabetes Obes Metab 2009;11:480-490. 25. Kamada H, Yu F, Nito C, et al. Influence of hyperglycemia on oxidative stress and matrix metalloproteinase-9 activation after focal cerebral ischemia/reperfusion in rats: relation to blood-brain barrier dysfunction. Stroke 2007; 38:1044-1049. 26. Liu J, Gao BB, Clermont AC, et al. Hyperglycemiainduced cerebral hematoma expansion is mediated by plasma kallikrein. Nat Med 2011;17:206-210. 27. Esposito KI, Nappo F, Marfella R, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 2002;106:2067-2072. 28. Martini SR, Kent TA. Hyperglycemia in acute ischemic stroke: a vascular perspective. J Cereb Blood Flow Metab 2007;27:435-451. 29. Kruyt ND, Biessels GJ, DeVries JH, et al. Hyperglycemia in aneurysmal subarachnoid hemorrhage: a potentially modifiable risk factor for poor outcome. J Cereb Blood Flow Metab 2010;30:1577-1587.