Computational pathology: A survey review and the way forward

Computational Pathology (CPath) is an interdisciplinary science that augments developments of computational approaches to analyze and model medical histopathology images. The main objective for CPath is to develop infrastructure and workflows of digital diagnostics as an assistive CAD system for clinical pathology, facilitating transformational changes in the diagnosis and treatment of cancer that are mainly address by CPath tools. With evergrowing developments in deep learning and computer vision algorithms, and the ease of the data flow from digital pathology, currently CPath is witnessing a paradigm shift. Despite the sheer volume of engineering and scientific works being introduced for cancer image analysis, there is still a considerable gap of adopting and integrating these algorithms in clinical practice. This raises a significant question regarding the direction and trends that are undertaken in CPath. In this article we provide a comprehensive review of more than 800 papers to address the challenges faced in problem design all-the-way to the application and implementation viewpoints. We have catalogued each paper into a model-card by examining the key works and challenges faced to layout the current landscape in CPath. We hope this helps the community to locate relevant works and facilitate understanding of the field's future directions. In a nutshell, we oversee the CPath developments in cycle of stages which are required to be cohesively linked together to address the challenges associated with such multidisciplinary science. We overview this cycle from different perspectives of data-centric, model-centric, and application-centric problems. We finally sketch remaining challenges and provide directions for future technical developments and clinical integration of CPath. For updated information on this survey review paper and accessing to the original model cards repository, please refer to GitHub. Updated version of this draft can also be found from arXiv.

Magnetite nanoparticles as a kinetically favorable source of iron to enhance GBM response to chemoradiosensitization with pharmacological ascorbate

Ferumoxytol (FMX) is an FDA-approved magnetite (Fe3O4) nanoparticle used to treat iron deficiency anemia that can also be used as an MR imaging agent in patients that can't receive gadolinium. Pharmacological ascorbate (P-AscH-; IV delivery; plasma levels ≈ 20 mM) has shown promise as an adjuvant to standard of care chemo-radiotherapy in glioblastoma (GBM). Since ascorbate toxicity mediated by H2O2 is enhanced by Fe redox cycling, the current study determined if ascorbate catalyzed the release of ferrous iron (Fe2+) from FMX for enhancing GBM responses to chemo-radiotherapy. Ascorbate interacted with Fe3O4 in FMX to produce redox-active Fe2+ while simultaneously generating increased H2O2 fluxes, that selectively enhanced GBM cell killing (relative to normal human astrocytes) as opposed to a more catalytically active Fe complex (EDTA-Fe3+) in an H2O2 - dependent manner. In vivo, FMX was able to improve GBM xenograft tumor control when combined with pharmacological ascorbate and chemoradiation in U251 tumors that were unresponsive to pharmacological ascorbate therapy. These data support the hypothesis that FMX combined with P-AscH- represents a novel combined modality therapeutic approach to enhance cancer cell selective chemoradiosentization in the management of glioblastoma.

Causes of Death in Endovascularly Treated Patients with Acute Stroke

BACKGROUND AND PURPOSE

Because stroke therapy has changed with the introduction of endovascular stroke treatment as a standard approach, studies on intrahospital causes of death from stroke are no longer up-to-date. The purpose of this observational study was to present the causes of death during hospitalization of patients with ischemic stroke who received endovascular stroke treatment, with the focus on a differentiation of curative and secondary palliative treatment.

MATERIALS AND METHODS

We studied a total cohort of 1342 patients who received endovascular stroke treatment in a tertiary stroke center (Aachen, Germany) between 2010 and 2020 and analyzed the causes of death in all 326 consecutive deceased patients. We distinguished between curative treatment and a secondary palliative approach and analyzed causes of death and treatment numbers across the years.

RESULTS

In the entire cohort of 326 deceased patients, the most common cause of death was of a cerebrovascular nature (51.5%), followed by pneumonia and sepsis (25.8%) and cardiovascular causes (8.3%). Neurovascular causes constituted 75.8% of reasons for palliation. In the group with a secondary palliative approach, causes of death were neurovascular in 54.0% of patients and pneumonia and sepsis in 26.0% of patients.

CONCLUSIONS

Cerebrovascular causes in patients with stroke play a major role in the intrahospital causes of death and reasons for palliation. Considering the large proportion of secondarily palliative-treated patients, reasons for palliation should be considered instead of causes of death to avoid concealment by, for example, life-terminating measures.

Frequency, Awareness, and Symptoms of Chikungunya Among Patients in a Tertiary Care Hospital of Karachi: A Cross-Sectional Study

Background 

Chikungunya fever (CHIKF) is an infectious illness spread by the bite of mosquitoes and caused by an arbovirus known as Chikungunya virus (CHIKV). This disease has become an epidemic in Karachi and various other cities of Pakistan, affecting a large population, especially those from poor, socioeconomically underdeveloped areas. It is proving to be a severe and alarming cause of debility due to its prolonged detrimental effects on the joints. A significant number of cases are reported daily in different hospitals of Karachi, with Civil Hospital being one of the major tertiary care hospitals. The aim of this study is to determine the frequency and symptoms of chikungunya as well as to assess the participants’ awareness about the spread and preventive measures of this disease.

Methods

This is a cross-sectional study that was carried out in Civil Hospital Karachi by approaching patients in the out-patient department (OPD) and the emergency department with complaints of fever and joint pain. All the data was collected via a pre-coded questionnaire during May-June 2017 by taking prior informed verbal consent and were analyzed through Statistical Package for the Social Sciences (SPSS) version 22 software.

Results

The age group most affected by this disease was the 21-30 years range, which represented almost a quarter of the cases (n=83, 32.17%). Majority of the respondents (n=214, 82.95%) had heard of the disease, mainly due to their own prior experience with it (n=100, 38.76%). Lethargy (n=219, 84.88%), difficulty in walking (n=213, 82.56%), and headache (n=209, 81.01%) were the major symptoms reported apart from fever (n=258, 100.00%) and arthralgia (n=258, 100.00%).

Conclusion

CHIKF is proving to be a great threat to people as it impairs their quality of life to a great extent. The recent outbreak of chikungunya has victimized a considerable population of Karachi. This study mainly assessed the severity of the disease and its symptoms as well as the lack of awareness among patients. Proper and effective preventive measures can further help to eradicate this disease on a large scale and prevent future epidemics.

Detection of optimal PEEP for equal distribution of tidal volume by volumetric capnography and electrical impedance tomography during decreasing levels of PEEP in post cardiac-surgery patients

Background

Homogeneous ventilation is important for prevention of ventilator-induced lung injury. Electrical impedance tomography (EIT) has been used to identify optimal PEEP by detection of homogenous ventilation in non-dependent and dependent lung regions. We aimed to compare the ability of volumetric capnography and EIT in detecting homogenous ventilation between these lung regions.

Methods

Fifteen mechanically-ventilated patients after cardiac surgery were studied. Ventilator settings were adjusted to volume-controlled mode with a fixed tidal volume (Vt) of 6–8 ml kg⁻¹ predicted body weight. Different PEEP levels were applied (14 to 0 cm H₂O, in steps of 2 cm H₂O) and blood gases, Vcap and EIT were measured.

Results

Tidal impedance variation of the non-dependent region was highest at 6 cm H₂O PEEP, and decreased significantly at 14 cm H₂O PEEP indicating decrease in the fraction of Vt in this region. At 12 cm H₂O PEEP, homogenous ventilation was seen between both lung regions. Bohr and Enghoff dead space calculations decreased from a PEEP of 10 cm H₂O. Alveolar dead space divided by alveolar Vt decreased at PEEP levels ≤6 cm H₂O. The normalized slope of phase III significantly changed at PEEP levels ≤4 cm H₂O. Airway dead space was higher at higher PEEP levels and decreased at the lower PEEP levels.

Conclusions

In postoperative cardiac patients, calculated dead space agreed well with EIT to detect the optimal PEEP for an equal distribution of inspired volume, amongst non-dependent and dependent lung regions. Airway dead space reduces at decreasing PEEP levels.

Lung stress and strain calculations in mechanically ventilated patients in the intensive care unit

Background

Stress and strain are parameters to describe respiratory mechanics during mechanical ventilation. Calculations of stress require invasive and difficult to perform esophageal pressure measurements. The hypothesis of the present study was: Can lung stress be reliably calculated based on non‐invasive lung volume measurements, during a decremental Positive end‐expiratory pressure (PEEP) trial in mechanically ventilated patients with different diseases?

Methods

Data of 26 pressure‐controlled ventilated patients admitted to the ICU with different lung conditions were retrospectively analyzed: 11 coronary artery bypass graft (CABG), 9 neurology, and 6 lung disorders. During a decremental PEEP trial (from 15 to 0 cmH₂O in three steps) end‐expiratory lung volume (EELV) measurements were performed at each PEEP step, without interruption of mechanical ventilation. Strain, specific elastance, and stress were calculated for each PEEP level. Elastance was calculated as delta PEEP divided by delta PEEP volume, whereas specific elastance is elastance times the FRC. Stress was calculated as specific elastance times the strain. Global strain was divided into dynamic (tidal volume) and static (PEEP) strain.

Results

Strain calculations based on FRC showed mainly changes in static component, whereas calculations based on EELV showed changes in both the static and dynamic component of strain. Stress calculated from EELV measurements was 24.0 ± 2.7 and 13.1 ± 3.8 cmH₂O in the lung disorder group at 15 and 5 cmH₂O PEEP. For the normal lungs, the stress values were 19.2 ± 3.2 and 10.9 ± 3.3 cmH₂O, respectively. These values are comparable to earlier publications. Specific elastance calculations were comparable in patients with neurologic and lung disorders, and lower in the CABG group due to recruitment in this latter group.

Conclusion

Stress and strain can reliably be calculated at the bedside based on non‐invasive EELV measurements during a decremental PEEP trial in patients with different diseases.

Pipeline Embolization Device in the Treatment of Recurrent Previously Stented Cerebral Aneurysms

BACKGROUND AND PURPOSE

The use of the Pipeline Embolization Device in the management of recurrent previously stented cerebral aneurysms is controversial. The aim of this study was to evaluate the efficacy and safety of the Pipeline Embolization Device in the treatment of recurrent, previously stented aneurysms.

MATERIALS AND METHODS

Twenty-one patients with previously stented recurrent aneurysms who later underwent Pipeline Embolization Device placement (group 1) were retrospectively identified and compared with 63 patients who had treatment with the Pipeline Embolization Device with no prior stent placement (group 2). Occlusion at the latest follow-up angiogram, recurrence and retreatment rates, clinical outcome, complications, and morbidity and mortality observed after treatment with the Pipeline Embolization Device were analyzed.

RESULTS

Patient characteristics were similar between the 2 groups. The mean time from stent placement to recurrence was 25 months. Pipeline Embolization Device treatment resulted in complete aneurysm occlusion in 55.6% of patients in group 1 versus 80.4% of patients in group 2 (P = .036). The retreatment rate in group 1 was 11.1% versus 7.1% in group 2 (P = .62). The rate of good clinical outcome at the latest follow-up in group 1 was 81% versus 93.2% in group 2 (P = .1). Complications were observed in 14.3% of patients in group 1 and 9.5% of patients in group 2 (P = .684).

CONCLUSIONS

The use of the Pipeline Embolization Device in the management of previously stented aneurysms is less effective than the use of this device in nonstented aneurysms. Prior stent placement can worsen the safety and efficacy profile of this device.

A single pipeline embolization device is sufficient for treatment of intracranial aneurysms

BACKGROUND AND PURPOSE

The Pipeline Embolization Device has emerged as an important treatment option for intracranial aneurysms. The number of devices needed to treat an aneurysm is uncertain and is the subject of vigorous debate. The purpose of this study was to compare rates of complications, aneurysm occlusion, and outcome in patients treated with a single-versus-multiple Pipeline Embolization Devices.

MATERIAL AND METHODS

One hundred seventy-eight patients were treated with the Pipeline Embolization Device at our institution. Patients were divided into 2 groups: a single-device group (n = 126) and a multiple-device group (n = 52).

RESULTS

There was no statistically significant difference between the 2 groups with respect to baseline characteristics except for age and aneurysm size (higher with multiple Pipeline Embolization Devices). Complications occurred more frequently with multiple (15%) versus a single device (5%, P = .03). In multivariate analysis, the use of multiple devices independently predicted complications. A similar proportion of patients achieved adequate aneurysm obliteration at follow-up in the single-device (84%) and the multiple-device groups (87%, P = .8). In multivariate analysis, age and follow-up time predicted obliteration. At follow-up, a significantly higher proportion of patients treated with a single device (97%) achieved a favorable outcome compared with those treated with multiple devices (89%, P = .03). In multivariate analysis, there was a strong trend for the use of a single device to predict favorable outcomes (P = .06).

CONCLUSIONS

Treatment with a single Pipeline Embolization Device provides similar occlusion rates with less complications and better overall outcomes. These findings suggest that a single Pipeline Embolization Device is sufficient for treatment of most intracranial aneurysms.

Coiling of large and giant aneurysms: complications and long-term results of 334 cases

BACKGROUND AND PURPOSE

Large and giant intracranial aneurysms are increasingly treated with endovascular techniques. The goal of this study was to retrospectively analyze the complications and long-term results of coiling in large and giant aneurysms (≥ 10 mm) and identify predictors of outcome.

MATERIALS AND METHODS

A total of 334 large or giant aneurysms (≥ 10 mm) were coiled in our institution between 2004 and 2011. Medical charts and imaging studies were reviewed to determine baseline characteristics, procedural complications, and clinical/angiographic outcomes. Aneurysm size was 15 mm on average. Two hundred twenty-five aneurysms were treated with conventional coiling; 88, with stent-assisted coiling; 14, with parent vessel occlusion; and 7, with balloon-assisted coiling.

RESULTS

Complications occurred in 10.5% of patients, with 1 death (0.3%). Aneurysm location and ruptured aneurysms predicted complications. Angiographic follow-up was available for 84% of patients at 25.4 months on average. Recanalization and retreatment rates were 39% and 33%, respectively. Larger aneurysm size, increasing follow-up time, conventional coiling, and aneurysm location predicted both recurrence and retreatment. The annual rebleeding rate was 1.9%. Larger aneurysm size, increasing follow-up time, and aneurysm location predicted new or recurrent hemorrhage. Favorable outcomes occurred in 92% of patients. Larger aneurysm size, poor Hunt and Hess grades, and new or recurrent hemorrhage predicted poor outcome.

CONCLUSIONS

Coiling of large and giant aneurysms has a reasonable safety profile with good clinical outcomes, but aneurysm reopening remains very common. Stent-assisted coiling has lower recurrence, retreatment, and new or recurrent hemorrhage rates with no additional morbidity compared with conventional coiling. Aneurysm size was a major determinant of recanalization, retreatment, new or recurrent hemorrhage, and poor outcome.

Spontaneous delayed migration/shortening of the pipeline embolization device: report of 5 cases

Five patients were found to have spontaneous delayed migration/shortening of their Pipeline Embolization Devices on follow-up angiography. The device migrated proximally in 4 patients and distally in 1 patient. One patient had a subarachnoid hemorrhage and died as a result of migration of the Pipeline Embolization Device, and another patient presented with complete MCA occlusion and was left severely disabled. Mismatch in arterial diameter between inflow and outflow vessels was a constant finding. Migration of the Pipeline Embolization Device was managed conservatively, with additional placement of the device, or with parent vessel occlusion. Obtaining complete expansion of the embolization device by using a longer device, increasing vessel coverage, using adjunctive aneurysm coiling, and avoiding dragging and stretching of the device are important preventive measures. Neurointerventionalists should be aware of this potentially fatal complication and take all necessary preventive measures.

The efficacy of endovascular stenting in the treatment of supraclinoid internal carotid artery blister aneurysms using a stent-in-stent technique

BACKGROUND AND PURPOSE

Blister aneurysms of the supraclinoid ICA represent a rare but well-documented cause of subarachnoid hemorrhage. These aneurysms are difficult to detect, and their surgical treatment is challenging, with high morbidity and mortality rates. The reports currently in the literature that describe the surgical and endovascular treatment of these aneurysms offer no clear consensus on the optimal treatment. We describe a staged endovascular treatment entailing stenting using a stent-in-stent technique, as well as planned but delayed embolization as the aneurysm increases in size to allow the introduction of coils.

MATERIALS AND METHODS

We performed a retrospective review of all cerebral angiograms performed at our institution over an 8-month period for evaluation of subarachnoid hemorrhage, identifying 6 ICA blister aneurysms.

RESULTS

All 6 blister aneurysms were located in the supraclinoid ICA. The stent-in-stent technique was used for the initial treatment of all patients. Three patients had no residual or recurrent aneurysm following initial treatment. Three patients required retreatment with coils after continued growth of the aneurysm, identified on follow-up angiography. Five patients had good recovery (average mRS score of 1), and 1 patient had poor neurologic recovery (mRS score of 3) due to a large hemorrhagic infarction.

CONCLUSIONS

Our case series suggests that staged endovascular treatment entailing the use of a stent-in-stent technique, augmented with subsequent coil embolization as necessary for progressive disease, is a viable endovascular option for treating ruptured supraclinoid blister aneurysms, allowing for parent artery preservation.

Utility of CT angiography in the identification and characterization of supraclinoid internal carotid artery blister aneurysms

BACKGROUND AND PURPOSE

Blister aneurysms of the supraclinoid ICA represent a rare but potentially catastrophic cause of SAH, often presenting both diagnostic and therapeutic dilemmas. We explore the utility of CTA in the identification and characterization of ICA blister aneurysms.

MATERIALS AND METHODS

We performed a retrospective review of catheter cerebral angiograms obtained at our institution over a 12-month period for evaluation of SAH, identifying 6 cases of ICA blister aneurysms. All patients underwent CTA and DSA for evaluation of SAH. The reports from the CTA and DSA studies were reviewed to identify aneurysms correctly diagnosed prospectively. Retrospective review of the CTA and DSA images was also performed. Review of the interpretations and images was performed for any follow-up studies.

RESULTS

All 6 patients presented with SAH, diagnosed by head CT. All patients subsequently underwent CTA prior to DSA evaluation. All 6 aneurysms were identified prospectively on initial DSA imaging. Of the 6 blister aneurysms, 4 (67%) were identified prospectively; and 5 (83%), retrospectively on CTA. All 6 patients underwent endovascular treatment with stent placement. Four of the 6 aneurysms underwent follow-up CTA (range, 9-22 days), including the 2 aneurysms that had been unidentifiable preprocedurally. All 4 blister aneurysms were seen postprocedurally by DSA. Three of these 4 (75%) residual aneurysms were detected by CTA (both prospectively and retrospectively).

CONCLUSIONS

In the presence of SAH and otherwise negative findings on CTA, a catheter cerebral angiogram should be performed to absolutely exclude an ICA blister aneurysm.

Impact of cardiac complications on outcome after aneurysmal subarachnoid hemorrhage: a meta-analysis

Impact of cardiac complications after aneurysmal subarachnoid hemorrhage (SAH) remains controversial. We performed a meta-analysis to assess whether EKG changes, myocardial damage, or echocardiographic wall motion abnormalities (WMAs) are related to death, poor outcome (death or dependency), or delayed cerebral ischemia (DCI) after SAH.

METHODS

Articles on cardiac abnormalities after aneurysmal SAH that met predefined criteria and were published between 1960 and 2007 were retrieved. We assessed the quality of reports and extracted data on patient characteristics, cardiac abnormalities, and outcome measurements. Poor outcome was defined as death or dependence by the Glasgow Outcome Scale (dichotomized at < or = 3) or the modified Rankin scale (dichotomized at > 3). If studies used another dichotomy or another outcome scale, we used the numbers of patients with poor outcome provided by the authors. We calculated pooled relative risks (RRs) with corresponding 95% confidence intervals for the relation between cardiac abnormalities and outcome measurements.

RESULTS

We included 25 studies (16 prospective) with a total of 2,690 patients (mean age 53 years; 35% men). Mortality was associated with WMAs (RR 1.9), elevated troponin (RR 2.0) and brain natriuretic peptide (BNP) levels (RR 11.1), tachycardia (RR 3.9), Q waves (RR 2.9), ST-segment depression (RR 2.1), T-wave abnormalities (RR 1.8), and bradycardia (RR 0.6). Poor outcome was associated with elevated troponin (RR 2.3) and creatine kinase MB (CK-MB) levels (RR 2.3) and ST-segment depression (RR 2.4). Occurrence of DCI was associated with WMAs (RR 2.1), elevated troponin (RR 3.2), CK-MB (RR 2.9), and BNP levels (RR 4.5), and ST-segment depression (RR 2.4). All RRs were significant.

CONCLUSION

Markers for cardiac damage and dysfunction are associated with an increased risk of death, poor outcome, and delayed cerebral ischemia after subarachnoid hemorrhage. Future research should establish whether these cardiac abnormalities are independent prognosticators and should be directed toward pathophysiologic mechanisms and potential treatment options.

Impact of early surgery after aneurysmal subarachnoid haemorrhage

OBJECTIVES

To investigate the effect of early aneurysm surgery (<72 h) on outcome in patients with subarachnoid haemorrhage (SAH).

MATERIALS AND METHODS

We studied two consecutive series of patients with aneurysmal SAH [postponed surgery (PS) cohort, n = 118, 1989-1992: surgery was planned on day 12 and early surgery (ES) cohort, n = 85, 1996-1998: ES was performed only in patients with Glasgow Coma Scale (GCS) >13]. We used multivariable logistic regression analysis to assess outcome at 3 months.

RESULTS

Favourable outcome (Glasgow Outcome Scale 4 or 5) was similar in both cohorts. Cerebral ischemia occurred significantly more often in the ES cohort. The occurrence of rebleeds was similar in both cohorts. External cerebrospinal fluid (CSF) drainage was performed more often in the ES cohort (51% vs 19%). Patients with cisternal sum score (CSS) of subarachnoid blood <15 on admission [adjusted odds ratio (OR) for favourable outcome: 6.4, 95% confidence interval (CI) 1.0-39.8] and patients with both CSS <15 and GCS > 12 on admission benefited from the strategy including ES (OR 10.5, 95% CI 1.1-99.4).

CONCLUSIONS

Our results support the widely adopted practice of ES in good-grade SAH patients.

Rationale of the SEAS study: Serial Echocardiography After Subarachnoid haemorrhage, a prospective national, multicentre, multidisciplinary, cohort study to evaluate cardiac abnormalities following intracranial haemorrhage

The SEAS study is a prospective national, multicentre, multidisciplinary, cohort study in which the cardiac abnormalities following aneurysmal subarachnoid haemorrhage are studied. Incidence, clinical implications and predictive variables of cardiac abnormalities following aneurysmal subarachnoid haemorrhage will be studied. Cardiac abnormalities are defined as ECG changes, echocardiographic function abnormalities, and biochemical changes. A total of 350 patients will be included over a period of three years including follow-up.

Effects of cerebral air embolism on brain metabolism in pigs

OBJECTIVES

Cerebral air embolism was induced in pigs and changes in intracranial pressure (ICP), brain oxygen (PbrO2), brain carbon dioxide (PbrCO2), brain pH (brpH) and glucose, lactate and pyruvate levels were used to characterize this model.

METHODS

In seven anesthetized pigs, ICP, PbrO2, PbrCO2 and brpH were measured continuously with multiparameter sensors and brain glucose metabolism by microdialysis. After injection of air into the internal carotid artery, these parameters were recorded for 2 h.

RESULTS

ICP increased (433%) from 12 +/- 1 to 52 +/- 8 mmHg (P < 0.05). PbrO2 decreased from 25.7 +/- 6.2 to 11.9 +/- 5.2 mmHg. PbrCO2 increased (109%) from 57.7 +/- 2.7 to 120.4 +/- 21.5 mmHg (P < 0.05). Brain glucose decreased (38%) from 3.05 +/- 0.91 to 1.91 +/- 0.55 mmol, while brain lactate increased (384%) from 1.36 +/- 0.15 to 5.22 +/- 0.53 mmol/l (P < 0.05).

CONCLUSIONS

Cerebral air embolism has a deleterious effect on ICP and brain metabolism. Therefore, this model may be suitable for testing therapeutic regimens in cerebral air embolism.

The clinical significance of whole blood viscosity in (cardio)vascular medicine

Whole blood is a non-Newtonian fluid, which means that its viscosity depends on shear rate. At low shear, blood cells aggregate, which induces a sharp increase in viscosity, whereas at higher shear blood cells disaggregate, deform and align in the direction of flow. Other important determinants of blood viscosity are the haematocrit, the presence of macro-molecules in the medium, temperature and, especially at high shear, the deformability of red blood cells. At the sites of severe atherosclerotic obstructions or at vasospastic locations, when change of vessel diameter is limited, blood viscosity contributes to stenotic resistance thereby jeopardising tissue perfusion. However, blood viscosity plays its most important role in the microcirculation where it contributes significantly to peripheral resistance and may cause sludging in the postcapillary venules. Apart from the direct haemodynamic significance, an increase in blood viscosity at low shear by red blood cell aggregation is also associated with increased thrombotic risk, as has been demonstrated in atrial fibrillation. Furthermore, as increased red blood cell aggregation is a reflection of inflammation, hyperviscosity has been shown to be a marker of inflammatory activity. Thus, because of its potential role in haemodynamics, thrombosis and inflammation, determination of whole blood viscosity could provide useful information for diagnostics and therapy of (cardio)vascular disease.

Diagnosis of perimesencephalic nonaneurysmal subarachnoid hemorrhage with computed tomography

OBJECTIVE

In 4% to 31% of patients with acute subarachnoid hemorrhage (SAH), no underlying cause is identified. Blood is restricted to the perimesencephalic cisterns in about two thirds of these patients. These patients are identified as having perimesencephalic nonaneurysmal subarachnoid hemorrhage (PNSAH), a syndrome based on the interpretation of computed tomography (CT) findings on admission, with an excellent prognosis, far better than other patients with SAH with or without an aneurysm. However, the diagnosis is subject to interrater variability, and differentiation between PNSAH and ruptured supratentorial aneurysm by means of CT has not been investigated. Therefore we investigated the validity of prediction of PNSAH with CT scan.

METHODS

A total of 303 consecutive patients, admitted within 72 hours after the initial bleeding with symptoms and signs of spontaneous SAH, which was confirmed by CT or spectrophotometric analysis of the cerebrospinal fluid (CSF) in whom either a selective 4-vessel angiogram (n = 293) or postmortem examination (n = 10) was performed, were studied.

RESULTS

We found an interrater agreement for the diagnosis perimesencephalic nonaneurysmal subarachnoid hemorrhage of 93% and a kappa value of 0.65. Sensitivity was 30% for observer 1 and 46% for observer 2. The positive predictive value was 73% for observer 1 and 76% for observer 2. Among those diagnosed as having PNSAH, we found vertebrobasilar artery aneurysms in 19% (both raters) and internal carotid artery aneurysms in 5% (observer 1) and 8% (observer 2).

CONCLUSION

We conclude that PNSAH can be distinguished on CT in the majority of patients; however, the angiographical management in PNSAH should not differ from other SAHs.

Cardioprotection in pigs by exogenous norepinephrine but not by cerebral ischemia-induced release of endogenous norepinephrine

BACKGROUND AND PURPOSE

Endogenous norepinephrine release induced by cerebral ischemia may lead to small areas of necrosis in normal hearts. Conversely, norepinephrine may be one of the mediators that limit myocardial infarct size by ischemic preconditioning. Because brief ischemia in kidneys or skeletal muscle limits infarct size produced by coronary artery occlusion, we investigated whether cardiac norepinephrine release during transient cerebral ischemia also elicits remote myocardial preconditioning.

METHODS

Forty-one crossbred pigs of either sex were assigned to 1 of 7 experimental groups, of which in 6 groups myocardial infarct size was determined after a 60-minute coronary occlusion and 120 minutes of reperfusion. One group served as control (no pretreatment), while the other groups were pretreated with either cerebral ischemia or an intracoronary infusion of norepinephrine.

RESULTS

In 10 anesthetized control pigs, infarct size was 84+/-3% (mean+/-SEM) of the area at risk after a 60-minute coronary occlusion and 120 minutes of reperfusion. Intracoronary infusion of 0.03 nmol/kg. min(-)(1) norepinephrine for 10 minutes before coronary occlusion did not affect infarct size (80+/-3%; n=6), whereas infusion of 0.12 nmol/kg. min(-)(1) limited infarct size (65+/-2%; n=7; P:<0.05). Neither 10-minute (n=5) nor 30-minute (n=6) cerebral ischemia produced by elevation of intracranial pressure before coronary occlusion affected infarct size (83+/-4% and 82+/-3%, respectively). Myocardial interstitial norepinephrine levels tripled during cerebral ischemia and during low-dose norepinephrine but increased 10-fold during high-dose norepinephrine. Norepinephrine levels increased progressively up to 500-fold in the area at risk during the 60-minute coronary occlusion, independent of the pretreatment, while norepinephrine levels remained unchanged in adjacent nonischemic myocardium and arterial plasma.

CONCLUSIONS

Cerebral ischemia preceding a coronary occlusion did not modify infarct size, which is likely related to the modest increase in myocardial norepinephrine levels during cerebral ischemia. The infarct size limitation by high-dose exogenous norepinephrine is not associated with blunting of the ischemia-induced increase in myocardial interstitial norepinephrine levels.

Interobserver variability of cisternal blood on CT after aneurysmal subarachnoid hemorrhage

Interobserver variability in the prediction of delayed cerebral ischemia by means of blood on CT was investigated in 159 patients with aneurysmal subarachnoid hemorrhage, admitted within 72 hours after the bleed. The authors found considerable interobserver variability in the assessment of the amount of blood in the individual cisterns. A high sum score was an independent predictor for delayed cerebral ischemia only for rater 1 (rater 1: hazard ratio, 3.26; 95% confidence interval [CI], 1.14 to 7.75; rater 2: hazard ratio, 1.72; 95% CI, 0.72 to 4.09). The authors conclude that interobserver variability limits the predictive power of the amount of blood on CT for the occurrence of cerebral ischemia.

Compression of the visual pathway by anterior cerebral artery aneurysm

Visual failure is an uncommon presenting symptom of an intracranial aneurysm. It is even more uncommon in aneurysms arising from the anterior cerebral artery (ACA). We presented 2 patients with an aneurysm of the A1 segment of the anterior cerebral artery causing visual field defects. One patient presented with a complete homonymous hemianopia due to compression of the optic tract by a giant aneurysm of the proximal left A1 segment. The second patient had an almost complete unilateral anopia caused by compression of the optic nerve and chiasm by an aneurysm of the distal part of the A1 segment with a small chiasmatic hemorrhage and ventricular rupture.

Validity of prediction of the site of ruptured intracranial aneurysms with CT

OBJECTIVE

We studied the diagnostic power of blood distribution on CT (performed within 72 hours after the bleed) for the site of ruptured aneurysm in 168 consecutive patients with subarachnoid hemorrhage with either a single aneurysm or no aneurysm on the four-vessel angiogram or postmortem examination.

METHODS

A neurosurgeon and a neuroradiologist blind to the results of the angiography independently scored the distribution of blood on the CT and predicted the site of the ruptured aneurysm.

RESULTS

Overall agreement among raters was 52% and chance-adjusted agreement (kappa) was 0.42 (weighted kappa value 0.47). A parenchymal cerebral hematoma was an excellent predictor for the site of a ruptured aneurysm but was present in only a minority of cases (15%). The next most valid predictor was blood distribution on CT in patients with a ruptured anterior cerebral artery aneurysm or anterior communicating artery aneurysm (sensitivity 0.79, specificity 0.96, and positive predictive value 0.79 for rater 1; sensitivity 0.77, specificity 0.97, and positive predictive value 0.90 for rater 2). The validity of the predictive value of blood distribution on CT in patients with a ruptured aneurysm of the middle cerebral artery, internal carotid artery, or posterior circulation arteries was either inconsistent between raters or low.

CONCLUSION

With the exception of the presence of a parenchymal hematoma, the site of the ruptured aneurysm can be predicted by CT only in ruptured anterior cerebral artery or anterior communicating artery aneurysms.

Impact of medical treatment on the outcome of patients after aneurysmal subarachnoid hemorrhage

BACKGROUND AND PURPOSE

The rationale behind early aneurysm surgery in patients with subarachnoid hemorrhage (SAH) is the prevention of rebleeding as early as possible after SAH. In addition, by clipping the aneurysm as early as possible, one can apply treatment for cerebral ischemia more vigorously (induced hypertension) without the risk of rebleeding. Hypervolemic hemodilution is now a well-accepted treatment for delayed cerebral ischemia. We compared the prospectively collected clinical data and outcome of patients admitted to the intensive care unit in the period 1977 to 1982 with those of patients admitted in the period 1989 to 1992 to measure the effect of the change in medical management procedures on patients admitted in our hospital with SAH.

METHODS

We studied 348 patients admitted within 72 hours after aneurysmal SAH. Patients with negative angiography results and those in whom death appeared imminent on admission were excluded. The first group (group A) consisted of 176 consecutive patients admitted from 1977 through 1982. Maximum daily fluid intake was 1.5 to 2 L. Hyponatremia was treated with fluid restriction (<1 L/24 h). Antihypertensive treatment with diuretic agents was given if diastolic blood pressure was >110 mm Hg. Patients in the second group (172 consecutive patients; group B) were admitted from 1989 through 1992. Daily fluid intake was at least 3 L, unless cardiac failure occurred. Diuretic agents and antihypertensive medications were avoided. Cerebral ischemia was treated with vigorous plasma volume expansion under intermittent monitoring of pulmonary wedge pressure, cardiac output, and arterial blood pressure, aiming for a hematocrit of 0.29 to 0.33. Aneurysm surgery was planned for day 12.

RESULTS

Patients admitted in group B had less favorable characteristics for the development of cerebral ischemia and for good outcome when compared with patients in group A. Despite this, we found a significant decrease in the frequency of delayed cerebral ischemia in patients of group B treated with tranexamic acid (P=0.00005 by log rank test) and significantly improved outcomes among patients with delayed cerebral ischemia (P=0.006 by chi2 test) and among patients with deterioration from hydrocephalus (P=0.001 by chi2 test). This resulted in a significant improvement of the overall outcome of patients in group B when compared with those in group A (P=0.006 by chi2 test). The major cause of death in group B was rebleeding (P=0.011 by chi2 test).

CONCLUSIONS

We conclude that the outcome in our patients with aneurysmal SAH was improved but that rebleeding remains a major cause of death. Patient outcome can be further improved if we can increase the efficacy of preventive measures against rebleeding by performing early aneurysm surgery.

Regional cerebral blood flow and cerebrovascular risk factors in the elderly population

Regional cerebral blood flow (rCBF) was studied in 60 elderly persons (aged 65 to 84 years) recruited from a population-based study, with single photon emission computed tomography using technetium 99m-labeled hexamethylpropylene amine oxime. We investigated whether it is only age that affects rCBF or whether other factors can be indentified that explain this relationship. Using multiple linear regression analysis, increasing age was significantly associated with rCBF decrease in parietal, temporo-parietal, and temporal cortex, but not in frontal cortex. Adjustment with several risk factors for cerebrovascular disease, including hypertension, history of myocardial infarction, factor VIIc, factor VIIIc, cholesterol and HDL cholesterol, smoking, and diabetes mellitus had no influence on these relations. Conversely, the association between age and rCBF was no longer statistically significant after adjustment with fibrinogen and indicators of carotid atherosclerosis, including intima-media wall thickness of the carotid artery and plaques in the carotid artery. Correction with local ratings of cortical atrophy did not affect the relations between age and rCBF. The results suggest that in the elderly population rCBF declines with age in posterior cortical areas and that these changes may well be explained by the presence of atherosclerosis. Reduced contractility of the vascular muscle wall with increasing age resulting from atherosclerosis may be the underlying mechanism.

Vascular risk factors, atherosclerosis, cerebral white matter lesions and cerebral perfusion in a population-based study

We studied risk factors for cerebral vascular disease (blood pressure and hypertension, factor VIIc, factor VIIIc, fibrinogen), indicators of atherosclerosis (intima-media thickness and plaques in the carotid artery) and cerebral white matter lesions in relation to regional cerebral blood flow (rCBF) in 60 persons (aged 65-85 years) recruited from a population-based study. rCBF was assessed with single-photon emission tomography using technetium-99m d, l-hexamethylpropylene amine oxime (99mTc-HMPAO). Statistical analysis was performed with multiple linear regression with adjustment for age, sex and ventricle-to-brain ratio. A significant positive association was found between systolic and diastolic blood pressure and temporo-parietal rCBF. In analysis with quartiles of the distribution, we found a threshold effect for the relation of low diastolic blood pressure (</=60 mmHg) and low temporo-parietal rCBF. Levels of plasma fibrinogen were inversely related to parietal rCBF, with a threshold effect of high fibrinogen levels (>3.2 g/l) and low rCBF. Increased atherosclerosis was related to low rCBF in all cortical regions, but these associations were not significant. No consistent relation was observed between severity of cerebral white matter lesions and rCBF. Our results may have implications for blood pressure control in the elderly population.

Outcome in patients with large intraventricular haemorrhages: a volumetric study

Patients with large amounts of intraventricular blood have a poor prognosis. The question is whether the quantity of ventricular blood or other factors related to the origin of the intraventricular bleeding cause poor outcome. The association between ventricular blood volume and outcome was analysed in patients with a primary intraventricular haemorrhage (PIVH) and in patients with an aneurysmal subarachnoid haemorrhage (SAH). The medical records of 24 patients with PIVH were reviewed and these patients were compared with 59 patients selected from a prospectively studied series of patients with aneurysmal SAH who had intraventricular blood without an intraparenchymal haematoma. Glasgow coma score on admission and outcome at three months expressed on the Glasgow outcome scale were available for all patients. Volumetric analysis on CT was carried out if there was at least a moderate amount of blood measured with a semi-quantitative scale (SAH n = 30, PIVH n = 22). In patients with SAH more than 20 ml intraventricular blood was lethal, as previously shown for patients with intraparenchymal haemorrhage. Patients with a PIVH can, however, survive such large volumes of intraventricular blood without impaired consciousness immediately after the bleeding, without developing hydrocephalus, and without any disability after three months. It is concluded that the cause of an intraventricular haemorrhage is more important for outcome than the amount of intraventricular blood, and large intraventricular haemorrhages are not always associated with hydrocephalus.

Detection of subarachnoid haemorrhage on early CT: is lumbar puncture still needed after a negative scan?

Computed tomography may be normal in up to 5% of patients who are investigated within one or two days after subarachnoid haemorrhage. This study investigated the need for further diagnostic evaluation after a normal CT scan was found very early (within 12 hours) in patients suspected of subarachnoid haemorrhage. A consecutive series of 175 patients with sudden headache and a normal neurological examination who had first CT within 12 hours after the onset of headache were investigated. The patients with normal CT underwent lumbar puncture, but not earlier than 12 hours after the event. Computed tomography showed subarachnoid blood in 117 patients, and was normal in 58. Spectrophotometric analysis of CSF gave evidence for a subarachnoid haemorrhage in two of these 58 patients (3%; 95% confidence interval (95% CI) 0.4-12%); a ruptured aneurysm was found in both. Thus CT was normal in two of 119 patients with a definite subarachnoid haemorrhage (2%; 95% CI 0.2-6%). It is concluded that in patients with sudden headache but normal CT a deferred lumbar puncture is necessary to rule out subarachnoid haemorrhage, even if CT is performed within 12 hours after the onset of symptoms.

Predictive factors for deterioration from hydrocephalus after subarachnoid hemorrhage

We studied the predictive factors for deterioration from hydrocephalus that developed during the first 28 days after admission in 660 patients following aneurysmal subarachnoid hemorrhage (SAH). Deterioration from hydrocephalus was defined as deterioration of consciousness with no detectable cause other than hydrocephalus confirmed by a repeat CT with a bicaudate index exceeding the 95th percentile for age. Deterioration from hydrocephalus occurred in 143 (22%) of the 660 patients. The variables included in the analysis were sex, age, loss of consciousness at ictus, sum score on the Glasgow Coma Scale on admission, sum score of cisternal blood and presence of ventricular blood on initial CT, hydrocephalus on initial CT, confirmed aneurysm, rebleeding, delayed cerebral ischemia, and treatment with tranexamic acid for 4 (short-term treatment) or 28 (long-term treatment) days. In a multivariate analysis with the Cox proportional hazards model incorporating fixed and time-dependent covariates, sum score of cisternal blood on initial CT (hazard ratio 3.15, p < 0.000001), presence of ventricular blood on initial CT (hazard ratio 1.66, p = 0.004), hydrocephalus on initial CT (hazard ratio 3.37, p < 0.000001), and long-term treatment with tranexamic acid (hazard ratio 2.40, p < 0.000001) were significantly related with the development of hydrocephalus. We conclude that a high amount of blood after SAH and delay of the resorption of cisternal and ventricular blood caused by long-term treatment with tranexamic acid increases the risk of deterioration from hydrocephalus after SAH.

Assessment of cerebral perfusion with single-photon emission tomography in normal subjects and in patients with Alzheimer's disease: effects of region of interest selection

The shape, size and location of regions of interest (ROIs) show considerable variability between single-photon emission tomography (SPET) studies in aging and Alzheimer's disease, but the possible influence on study results remains unknown. We compared three different ROIs in a SPET study with 60 controls and in 48 patients with probable Alzheimer's disease diagnosed according to the NINCDS-ADRDA criteria. Regional cerebral blood flow (rCBF) was assessed with SPET using technetium-99m d,l-hexamethylpropylene amine oxime (99mTc-HMPAO), normalized to the mean activity in a cerebellar reference slice. The three different ROIs were: a multi-slice and a single-slice ROI with reference to the normal brain anatomy (using an anatomical atlas), and a rectangular (2 x 4 pixels) ROI in the frontal, temporal, temporoparietal and occipital cortices. No differences were observed for the means of rCBF values between the single-slice and multi-slice ROI's with reference to the normal anatomy, but some variability was present for individual comparisons. In contrast, significantly higher mean rCBF values were obtained with the single-slice rectangular ROIs in all four regions for both patients and controls and considerable variability was shown for individual subjects. After analysis with multivariate logistic regression and receiver operator characteristic curves, the ability of SPET to discriminate between controls and Alzheimer patients was similar in the three methods for mild and moderate Alzheimer patients (Global Deterioration Scale = GDS of 3 and 4). However, with increasing dementia severity (GDS > 4) the rectangular ROIs showed lower ability to discriminate between groups compared to the single-slice and multi-slice anatomically defined ROIs.(ABSTRACT TRUNCATED AT 250 WORDS)

Definition of initial grading, specific events, and overall outcome in patients with aneurysmal subarachnoid hemorrhage. A survey

BACKGROUND AND PURPOSE

Scientific communication in medicine can be effective only if reports are based on unequivocal criteria for clinical conditions or specific diagnoses.

METHODS

We reviewed all articles about subarachnoid hemorrhage published in nine neurosurgical or neurological journals from 1985 through 1992 and assessed the presence and the precision of definitions used for reporting the initial grade, the specific complications of rebleeding, delayed cerebral ischemia, and hydrocephalus, and the overall outcome. We identified 184 articles reporting direct observations in at least 10 patients on one or more of these conditions.

RESULTS

Of 161 articles reporting the initial condition, only 19% used an unequivocal grading system (World Federation of Neurological Surgeons Scale or Glasgow Coma Scale); this proportion did not increase after 1988, when the World Federation of Neurological Surgeons Scale was introduced. The specific outcome events of rebleeding, ischemia, and hydrocephalus (283 instances) were sufficiently defined in only 31% of instances, incompletely in 22%, and not at all in 47%. The proportions were similar when the results were analyzed according to the type of complication, the year of publication, or per study. The four exclusively neurosurgical journals featured suitable definitions for any of the three outcome events in 20% of 209 instances, whereas the five mainly neurological journals published fewer articles about subarachnoid hemorrhage (74 instances of outcome events) but more often with precise criteria (65%). Overall outcome was adequately reported in 63% of all articles, with an increase over the years (54% in 1985 through 1988, 71% in 1989 through 1992).

CONCLUSIONS

Reports about subarachnoid hemorrhage require closer scrutiny before publication to ascertain whether the conclusions about specific outcome events are based on unequivocal criteria.

The diagnostic value of SPECT with Tc 99m HMPAO in Alzheimer's disease: a population-based study

We studied the diagnostic accuracy of single-photon emission computed tomography (SPECT) with technetium 99m-labeled hexamethylpropylene amine oxime (Tc 99m HMPAO) in 48 patients with probable Alzheimer's disease (AD) according to NINCDS-ADRDA criteria and in 60 controls recruited from a population-based study. With logistic regression, we identified decreased temporal regional cerebral blood flow as the best discriminating variable between patients and controls. Receiver-operator characteristic curves showed that the discriminative ability of SPECT improved with increasing dementia severity. With specificity set at 90%, sensitivity figures were 42% in mild, 56% in moderate, and 79% in severe AD. The diagnostic gain as a function of the prior probability of the disease being present was computed for those with mild AD. When the prior probability varied at around 50%, the diagnostic gain for mild AD patients was substantial (a maximum of 34%) for a positive test result but poor for a negative test result. The results suggest that the practical usefulness of SPECT as a diagnostic adjunct in patients suspected of having mild AD is confined to situations in which, on clinical grounds, there is considerable diagnostic doubt.

Amount of blood on computed tomography as an independent predictor after aneurysm rupture

BACKGROUND AND PURPOSE

After admission to the hospital of patients with aneurysmal subarachnoid hemorrhage, we assessed the predictive value of the extent of the hemorrhage on computed tomography in addition to that of clinical grading scales for poor outcome, infarction, and rebleeding.

METHODS

We studied 471 consecutive patients with aneurysmal subarachnoid hemorrhage and used logistic regression with step-wise forward selection of variables.

RESULTS

On admission, poor outcome was predicted by a low Glasgow Coma Scale score (odds ratio, 0.8; 95% confidence interval, 0.7-0.9); treatment with fluid restriction (2.5; 1.6-4.0); age over 52 (2.6; 1.7-3.9); loss of consciousness at ictus (1.7; 1.1-2.6); or a large amount of subarachnoid blood (2.0; 1.3-3.1). Delayed infarction was predicted by a large amount of subarachnoid blood (1.8; 1.2-2.6) or treatment with tranexamic acid (1.6; 1.1-2.4). Rebleeding was predicted by treatment with tranexamic acid (0.4; 0.3-0.7; protective effect); age over 52 (1.9; 1.2-3.0); loss of consciousness at ictus (1.7; 1.1-2.7); or admission to a neurosurgery service (0.6; 0.3-0.9; protective effect). Comparison of the observed and predicted outcome events showed that inclusion of the amount of subarachnoid blood into a predictive model added little to the prediction of poor outcome in general, but much to the prediction of delayed cerebral ischemia.

CONCLUSIONS

The total amount of subarachnoid blood on the initial computed tomogram has independent predictive power for the occurrence of delayed cerebral ischemia.

Epileptic seizures after subarachnoid hemorrhage

We studied predictive factors for the occurrence of epilepsy in 381 consecutive patients admitted within 72 hours after they had a subarachnoid hemorrhage from a ruptured intracranial aneurysm. Fits occurring in the presence of hyponatremia or within 12 hours after the initial bleed, rebleeding, or aneurysm surgery were classified as associated with these acute events and we did not regard these fits subsequent epileptic seizures. Thirty-five patients (9%) had one or more epileptic seizures, 12 hours to 1,761 days after the initial bleed (median value, 18 days). The following variables were included in the analysis: sex, age, history of hypertension, history of cardiovascular disease, loss of consciousness at ictus, sum score on the Glasgow Coma Scale on admission, sum score for the amount of cisternal blood and presence of intraventricular blood based on the initial computed tomography (CT) scan, occurrence of ictal seizures (seizures occurring within 12 hours after the onset), acute hydrocephalus, rebleeding, delayed cerebral ischemia, fluid intake, treatment with tranexamic acid, ventricular drainage, and aneurysm surgery. After multivariate analysis by means of Cox proportional hazards model with stepwise forward selection of the variables, a high cisternal blood score and rebleeding proved to be significantly related to epilepsy (hazard ratio = 2.06, p = 0.040; and hazard ratio = 3.02, p = 0.016), even after the exclusion of 28 patients who received perioperative prophylactic anticonvulsant therapy (hazard ratio = 2.31, p = 0.022; and hazard ratio = 3.65, p = 0.006, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Acute hydrocephalus in nonaneurysmal perimesencephalic hemorrhage: evidence of CSF block at the tentorial hiatus

The acute hydrocephalus in patients with nonaneurysmal perimesencephalic hemorrhage suggests an extraventricular obstruction of CSF flow. We studied the occurrence of acute hydrocephalus and the site of cisternal blood in 40 consecutive patients with perimesencephalic hemorrhage. In all 11 patients with hydrocephalus, all perimesencephalic cisterns were filled with blood; this occurred in only five of the 29 patients (17%) without hydrocephalus (p less than 0.0001). We conclude that in the absence of intraventricular blood, filling of all perimesencephalic cisterns with blood is a necessary factor for the development of acute hydrocephalus.

Distribution of cisternal blood in patients with acute hydrocephalus after subarachnoid hemorrhage

The distribution of cisternal blood in relation to the development of acute hydrocephalus was studied in 246 consecutive patients with aneurysmal subarachnoid hemorrhage who were admitted within 72 hours. Patients with evidence on the initial computed tomograph (CT) of subarachnoid hemorrhage caused by other than a ruptured aneurysm and patients with a negative angiography were excluded. Acute hydrocephalus (defined as a bicaudate index, measured on the initial CT or on a repeat CT within 1 week after subarachnoid hemorrhage, exceeding the 95th percentile for age) was found on the initial CT in 50 (20%) of the 246 patients and on a repeat CT in 9 other patients. Ventricular blood was found significantly more often in patients with acute hydrocephalus than in those in whom acute hydrocephalus did not develop (28 of 59 [47%] versus 58 of 187 [31%]; chi 2 = 4.634, p = 0.031). When the analysis was restricted to the 86 patients with ventricular blood, no significant differences were found in the total amount of cisternal blood and in the distribution of cisternal blood between patients with and without hydrocephalus. In contrast, among the 160 patients without ventricular blood, hydrocephalus was associated with a slightly higher total amount of cisternal blood (Wilcoxon's rank sum test, p = 0.023), and significantly more patients with acute hydrocephalus had a higher score in both ambient cisterns than patients without acute hydrocephalus (20 of 31 [65%] versus 41 of 129 [32%]; chi 2 = 10.007, p = 0.002).(ABSTRACT TRUNCATED AT 250 WORDS)

Outcome in patients with subarachnoid haemorrhage and negative angiography according to pattern of haemorrhage on computed tomography

15% of patients with spontaneous subarachnoid haemorrhage have normal cerebral angiograms; they fare better than patients with demonstrated aneurysms, though rebleeding and cerebral ischaemia can still occur. In patients with a normal angiogram and accumulation of blood in the cisterns around the midbrain--"perimesencephalic nonaneurysmal haemorrhage"--outcome is excellent. To test the hypothesis that rebleeding and disability in angiogram-negative subarachnoid haemorrhage might be limited to those with other patterns of haemorrhage on initial computed tomography (CT), complications and long-term outcome were studied in 113 patients with angiogram-negative subarachnoid haemorrhage, admitted between January, 1983, and July, 1990. All patients were investigated with third-generation CT scans within 72 h of the event, and with cerebral angiography. The mean follow-up period was 45 (range 6-96) months. None of 77 patients with a perimesencephalic pattern of haemorrhage on CT died or was left disabled as a result of the haemorrhage (0% [95% confidence interval 0-5%]). Among the other 36 patients, who had a blood distribution on CT indistinguishable from that in proven aneurysmal bleeds, 4 had rebleeds and 9 died or were left disabled as result of the haemorrhage (25% [14-43%]). Thus, two distinct subsets of patients with angiogram-negative subarachnoid haemorrhage should be recognised. Patients with a perimesencephalic pattern of haemorrhage have an excellent prognosis. Rebleeding, cerebral ischaemia, and residual disability occur exclusively in patients with aneurysmal patterns of haemorrhage on initial CT. Repeated angiography in search of an occult aneurysm is justified only in the patients with aneurysmal patterns.

Single photon emission computed tomography in patients with acute hydrocephalus or with cerebral ischaemia after subarachnoid haemorrhage

Using single photon emission computed tomography (SPECT), cerebral blood flow was studied in eight patients with gradual deterioration in the level of consciousness after subarachnoid haemorrhage. Four had cerebral ischaemia and four had acute hydrocephalus. In patients with cerebral ischaemia, single photon emission computed tomography scanning showed multiple regions with decreased uptake of technetium-99M labelled d,l-hexamethyl-propylene amine oxime (99mTcHM-PAO) mainly in watershed areas. In patients with acute hydrocephalus, decreased uptake was seen mainly in the basal parts of the brain: around the third ventricle, around the temporal horns of the lateral ventricles, and in the basal part of the frontal lobe. After serial lumbar puncture, there was improvement of the uptake of 99mTc HM-PAO in these basal areas in three (convincingly in two and slightly in the other) of the four patients accompanied by clinical improvement in these three patients. These results suggest that patients with acute hydrocephalus and impaired consciousness after SAH, in contrast to patients with cerebral ischaemia, have decreased cerebral blood flow predominantly in the basal parts of the brain.

The clinical course of perimesencephalic nonaneurysmal subarachnoid hemorrhage

We studied the early clinical course of 65 patients with perimesencephalic (nonaneurysmal) subarachnoid hemorrhage. None of the patients rebled; none had delayed cerebral ischemia; and only 3 patients (5%) developed clinical signs of acute hydrocephalus, 2 requiring ventricular shunting. Hyponatremia and electrocardiographic changes were found in the same proportions as in patients with aneurysmal rupture. All patients had a good outcome after 3 months. To control for the influence of the relatively small amount of cisternal blood in perimesencephalic hemorrhage on the absence of delayed cerebral ischemia, we compared these 65 patients with 49 patients who had aneurysmal subarachnoid hemorrhage. This control group with aneurysms was selected according to a similar amount of cisternal blood, a comparable level of consciousness on admission, and absence of other factors known to increase the risk for delayed cerebral ischemia. We found that 4 of the 49 patients with aneurysmal hemorrhage developed delayed cerebral ischemia (Fisher's exact test, p = 0.03); therefore the relatively small amount of blood does not account for the absence of delayed cerebral ischemia in perimesencephalic hemorrhage. Patients with a perimesencephalic pattern of hemorrhage and a normal angiogram should be considered to have a distinct subset of subarachnoid hemorrhage and should be excluded from future treatment trials of patients with subarachnoid hemorrhage.

Treatment of acute hydrocephalus after subarachnoid hemorrhage with serial lumbar puncture

Computed tomography demonstrated acute hydrocephalus less than or equal to 72 hours after subarachnoid hemorrhage in 24 (23%) of 104 patients. Of these 24 patients, six (25%) had no impairment of consciousness. In nine (11%) of the remaining 80 patients, acute hydrocephalus developed within 1 week after subarachnoid hemorrhage. With the exception of three patients, all 104 patients received antifibrinolytic treatment. Delayed clinical deterioration from acute hydrocephalus occurred in seven (29%) of the 24 patients with acute hydrocephalus on admission and in six (8%) of the remaining 80 patients. Serial lumbar puncture was performed in 17 patients. Twelve (71%) of the 17 patients treated with serial lumbar puncture, including 10 (77%) of the 13 patients with delayed deterioration from acute hydrocephalus after admission, achieved improvement in the level of consciousness. Four of these 17 patients (4% of all 104 patients) required an internal shunt. No patient deteriorated from coning following serial lumbar puncture. The rebleeding rate within 12 days after subarachnoid hemorrhage in hydrocephalic patients with serial lumbar puncture was not higher than the rate in those without hydrocephalus (two [12%] of 17 versus nine [13%] of 71). Neither meningitis nor ventriculitis was observed. We conclude that if neither a hematoma with a mass effect nor an obstructive element exists, cerebrospinal fluid drainage with serial lumbar puncture is a good alternative to ventricular drainage in patients with acute hydrocephalus after subarachnoid hemorrhage.