- Email: [email protected]
Factors Affecting Attenuation of Dural Sinuses on Noncontrasted Computed Tomography Scan
ARTICLE IN PRESS
Factors Affecting Attenuation of Dural Sinuses on Noncontrasted Computed Tomography Scan Nosaiba T. Al-Ryalat, MD (Doctor of Medicine),* Saif Aldeen S. AlRyalat, MS (Medical Student),† Lna W. Malkawi, MS,† Ethar F. Al-Zeena, MD,† Mahasen S. Al Najar, MD,* and Azmy M. Hadidy, MD, FCRC (Fellow of the Royal Collage of Radiologists)*
Background and Purpose: Noncontrasted computed tomography (NCCT) is used as the initial neuroimaging test of choice for patients who present with newonset neurological symptoms. An apparently hyperattenuated venous sinus may lead to the suspicion of cerebral venous sinus thrombosis (CVST). Improved understanding of all factors that can affect attenuation of dural sinuses can guide triage of patients to or from further investigations of suspected CVST. The purpose of this retrospective study was to assess the effect of different factors including hematocrit (HCT), hemoglobin (Hb), age, BUN/Cr ratio (blood urea nitrogen-tocreatinine ratio), and gender on the attenuation of dural sinuses on brain NCCT. Methods: A total of 1293 patients with neurological symptoms who presented to the emergency department were included in this study. For each patient, clinical assessment, laboratory investigations, and brain NCCT were reviewed. For each brain NCCT, the average attenuation of superior sagittal sinus and both right and left sigmoid sinuses was measured. Results: Positive significant correlations were found between average attenuation of dural sinuses on one hand and each of age, Hb, and HCT on the other hand. No significant correlation was found between average attenuation and BUN/Cr ratio. Gender discrepancy was also significant as higher attenuation was found in men. Conclusion: Age, gender, and Hb levels are the main factors that should be taken into account upon the assessment of dural sinuses on brain NCCT. The highest normal attenuation is predicted in an elderly polycythemic man and the lowest is predicted in a young anemic woman. Key Words: Cerebrovenous sinus thrombosis—noncontrasted CT—dural sinuses—attenuation—age. © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.
From the *Radiology and Nuclear Medicine Department, Jordan University Hospital, The University of Jordan, Amman, Jordan; and †Faculty of Medicine, Jordan University Hospital, The University of Jordan, Amman, Jordan. Received March 24, 2016; revision received May 18, 2016; accepted July 2, 2016. Address correspondence to Saif Aldeen S. AlRyalat, MS (Medical Student), Faculty of Medicine, Jordan University Hospital, The University of Jordan, P.O. Box 1669 Tela Al Ali, 11953 Amman, Jordan. E-mail: [email protected]. 1052-3057/$ - see front matter © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.07.002
Introduction Cerebral venous sinus thrombosis (CVST) is a challenging neurologic disorder that represents approximately 0.5% of all stroke cases worldwide.1 The symptoms and clinical features are nonspecific leading to a highly variable presentation among different age groups, making the diagnosis of CVST mainly based on a high index of clinical suspicion and imaging confirmation. Despite being a rare neurologic disorder, with an annual incidence estimated at 2-7 cases per 1 million populations,2 mortality rates of 10% have been reported among CVST patients,1 making early diagnosis and treatment very critical.
Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2016: pp ■■–■■
1
ARTICLE IN PRESS N.T. AL-RYALAT ET AL.
2
Noncontrasted computed tomography (NCCT) scan is widely used as the initial neuroimaging test of choice for patients who present to the emergency department with new-onset neurological symptoms such as headache, seizures, mental alteration, or focal neurological signs.3 As subjective assessment of the radiodensity based on apparent increase or decrease in attenuation may be misleading,2 radiologists tend routinely to measure the attenuation of intracerebral structures on brain NCCT by means of Hounsfield units (HU), where observing a dural sinus with an elevated HU value (i.e., hyperattenuation) may lead to the suspicion of CVST.4 These hyperattenuation signs have a crucial role in detecting CVST at an early stage when treatment is most likely to be effective and to greatly affect the clinical outcome.5 The drawback of these signs is that many factors can affect the degree of attenuation of blood, therefore confusing normality with pathologic conditions and thus increasing the number of false-positive and false-negative readings upon interpretation of NCCT. Improved understanding of all factors that can affect attenuation of dural sinuses can guide triage of patients to or from further investigations of suspected CVST. Despite the crucial role of NCCT, still minimal information is known about these factors. Hemoconcentration and dehydration have long been assumed as known factors to increase the attenuation of blood on NCCT,2,4,6 but the correlation with many other factors is not established yet. The purpose of this retrospective study, which included one of the largest samples of control population ever used for the same purpose, is to assess the effect of different factors including hematocrit (HCT), hemoglobin (Hb), age, BUN/Cr ratio (blood urea nitrogen-to-creatinine ratio), and gender on the attenuation of dural sinuses on brain NCCT.
Methodology This retrospective study was approved by our institutional review board; patient informed consent was waived.
Patients A total of 1580 patients who presented to the emergency department in our tertiary hospital in the period between June 28, 2014, and September 5, 2015, were recruited in this study; clinical assessment, laboratory investigations, and NCCT of brain were reviewed for each patient. To meet our inclusion criteria, (1) each patient was presented with new-onset neurological symptoms such as headache, seizure, mental alteration, or focal neurological signs; (2) NCCT scan, complete blood count (CBC), and kidney function test (KFT) were done within 24 hours of onset of symptoms; and (3) diagnosis of CVST was excluded by magnetic resonance venography (MRV), or diagnosis other than CVST was established. The following were the exclusion criteria: (1) clinically dehydrated patients as indicated in their hydration
assessment sheet; (2) hemorrhage or skull fracture adjacent to dural sinuses, severe brain swelling, intra-axial or extra-axial mass, contrast media administration within the previous 24 hours, or patients who received blood transfusion within the previous 24 hours; and (3) neonates below 6 months in order to eliminate the effect of fetal Hb. Up to 1293 patients were included in our study. Of these patients, 286 were excluded according to our criteria: 142 patients due to clinical dehydration; 93 due to head pathologies mentioned in criteria #2; and 51 due to age less than 6 months, with only 1 excluded due to CVST confirmed by MRV. Male–female difference was not significant as 671 (51.9%) were male and 622 (48.1%) were female. Of the 1293 patients included, 413 presented with various neurological signs and symptoms including headache, dizziness, nausea, and vomiting without focal neurological signs; they had normal CT and normal magnetic resonance venography later on (so CVST was excluded). Two hundred five patients had focal neurological signs and negative CT, had their MRI done, and showed an infarction; furthermore, other 162 patients had focal neurological signs and negative CT and MRI and diagnosed with transient ischemic attack. Up to 164 patients presented mainly with headache and diagnosed with sinusitis on CT. One hundred seventeen patients presented with neurological signs and symptoms with fever and negative CT and diagnosed with lumber puncture with meningitis. One hundred nine patients presented with headache and high blood pressure, had their brain CT done, and showed no brain hemorrhage. One hundred twenty-three patients presented with neurological symptoms and found to have other system-related infections.
Data Interpretation Attenuation of superior sagittal sinus and both left and right sigmoid sinuses was measured by 1 specialized radiologist in order to eliminate interobserver variation, 1 region of interest (ROI) measuring 2-mm recorder used to measure attenuation of each sinus in HU, except in small nondominant sigmoid sinuses in which we used an ROI of 1 mm. The average sinus density was calculated by taking the mean of density of superior sagittal sinus and both left and right sigmoid sinuses. Hb and HCT were obtained from CBC whereas Cr and BUN were obtained from the KFT. BUN/Cr ratio was calculated. The Hounsfield number of superior sagittal sinus to HCT level (H:H ratio, Hounsfield unit-tohematocrit ratio) was calculated.
Imaging Protocol All images were done via Siemens, Somatom Definition Flash 128 slice, CT scanner (Siemens Medical Solutions, Forchheim, Germany), with the following parameters: 360-410 mAs, 120 kV, section thickness of 5 mm, and reconstruction increment of 5 mm.
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
Statistical Analysis In our statistical analysis, we used IBM SPSS for Windows release 21.0 (SPSS Inc., Chicago, IL); descriptive statistics was used to describe our sample population. Correlation test was used to study the relation between age and HCT; partial correlation tests were used to study the relation between average attenuation and each of the following: age, HCT, Hb, and BUN/Cr ratio (by controlling other factors when studying each one); independentsample t-test was used to study the relation between average attenuation and gender, it was also used to compare BUN/Cr ratios above and below 20. Multivariate and univariate analyses were used to compare different variables including age, HCT, average attenuation, and BUN/Cr ratio with gender. A P value < .05 was taken as significant.
Results A total of 1293 patients from all ages were included with a mean age of 55.5 years (ranged from 1 to 97 years) without significant gender difference (mean age for men was 55.1 years and for women was 55.8 years).
HCT/Hb The mean HCT for our sample population was 39.4 (ranged from 13.6 to 58.9) whereas for Hb the mean was 12.9 (ranged from 6.3 to 18.7). Negative significant correlation (−.12) was found between age and HCT with a P value < .001; this correlation was also found to become weaker (−.07) but significant (P = .008) upon controlling Cr. Positive significant correlation (.48) was found between the average attenuation and HCT with a P value < .001; a stronger correlation was found between average attenuation and Hb (.50) with a P value < .001. Mean values of attenuation for anemic, normal, and polycythemic patients are mentioned in Table 1. To calculate the H:H ratio, we used the attenuation values of superior sagittal sinus in order to compare our ratio Table 1. (a) Male and (b) female densities HCT/age (a) Anemic Normal Polycythemic (b) Anemic Normal Polycythemic
<18
18-65
>65
49.6 53.5 55.4
55.0 55.5 56.2
52.5 55.0 61.4
47.9 50.9 53.7
50.7 53.2 55.6
51.8 52.9 59.7
Abbreviation: HCT, hematocrit. (a) Anemic: HCT < 40, normal: 40 < HCT < 54, polycythemic: HCT > 54. (b) Anemic: HCT < 36, normal: 36 < HCT < 46, polycythemic: HCT > 46.
3
with those in other studies; the maximum H:H ratio reached 3.82, where 31 patients had a ratio of more than 2.
Age Positive and significant correlation (.12) was found between age and average attenuation with a P value < .001; this correlation was found upon controlling the effect of HCT. The mean attenuation for different age groups for both men and women is shown in Table 1, a,b.
BUN/Cr Ratio The mean BUN/Cr ratio for our sample was 16.3. Correlation between average attenuation on one hand and each of BUN/Cr ratio, BUN, and Cr on the other hand was found to be insignificant with P values of .27, .55, and .43, respectively. No significant difference was found between attenuation of patients with BUN/Cr ratio above 20 and of those with a ratio below 20 (P value = .11).
Gender On a multivariate test, a significant P value < .001 was found. On Levene’s test of equality, the significance was only for HCT (P < .001) and average attenuation (P = .01); no significance was found for age or BUN/Cr ratio. Gender discrepancy was found to be significant regarding average attenuation using partial correlation test (with controlling the effect of HCT) with a P value = .02. The mean attenuation for men was 53.9 HU and for women was 52.4 HU (Table 1).
Discussion NCCT scan is performed as the first-line imaging modality in the emergency setting of most institutions for patients who present with new-onset neurological symptoms; this is the function of NCCT being a cost-effective, fast, and widely available imaging technique.2,7 The classic sign of acute CVST on NCCT is an increase in attenuation of the thrombosed venous sinus.2 Because a moderate increase in attenuation may not be always picked up by the radiologists, measurement of attenuation values of the sinuses by means of HU might be a more objective and reliable method. A recent study estimated the sensitivity of NCCT in the detection of CVST at 73%7; this may be even lower in daily clinical practice where multiple patientrelated factors can affect the attenuation of cerebral vasculature increasing possibility of false-positive or falsenegative interpretation of sinus attenuation. Where a prompt diagnosis of CVST is of critical importance as anticoagulation therapy is thought to prevent propagation of the thrombus,8 a false-positive diagnosis can, on the other hand, cause a delay in establishing the correct diagnosis as well as proceeding to inappropriate treatment.9 Using a large sample of controls with a widely variable range of attenuation values, we tried to assess
ARTICLE IN PRESS 4
the effect of many factors on attenuation of venous sinuses trying to reach the maximum possible sensitivity for NCCT in the diagnosis of CVST. Starting with HCT and Hb levels, a significant association between blood attenuation and both Hb and HCT levels has long been established.2,4,10 This correlation is also appreciated in our study which showed a positive and significant correlation between average density of dural sinuses and both Hb and HCT levels; a stronger linear relationship with Hb than that with HCT was demonstrated. It was reported that patients with polycythemia vera, who have high Hb and HCT levels, have hyperattenuated blood.11 Furthermore, blood of anemic patients, who have low Hb levels, was found to appear hypoattenuated on NCCT.12 Our study included both anemic and polycythemic patients in addition to patients with normal Hb; Table 1, a,b shows the mean sinus attenuation for each group. Black et al4 suggested the potential usefulness of HU values and H:H ratio in the diagnosis of CVST; they found that hemoconcentration correlates with attenuation of cerebral venous sinuses on NCCT. They were the first to use the H:H ratio to assess the correlation between hyperattenuation on NCCT and HCT level and recommended an H:H ratio greater than or equal to 2 as a cutoff value to separate patients with CVST from those without. Applying the H:H threshold suggested by Black et al to our study, which included a large sample of controls without CVST, 31 cases (2.5%) had an H:H ratio greater than or equal to 2 (ranging from 2.00 to 3.67), a large number that might be easily misinterpreted as cases of CVST increasing number of false-positive readings. A subsequent study suggested an H:H ratio greater than 1.52 to indicate a strong likelihood of a clot,2 a lower threshold that will result in a higher number of false-positive readings if HCT was the only factor taken into consideration upon interpreting a brain NCCT. Although they successfully controlled the effect of HCT factor by using the H:H ratio, other factors are still affecting the attenuation level giving rise to this large number of potentially false-positive readings; these factors also need to be corrected before reaching the final diagnosis. Noticeably, Black et al suggested an HU greater than 70 to be suspicious for CVST and to warrant further confirmatory investigations. Followed by Buyck et al,2 a more recent study suggested that in patients with an attenuation of more than 62 HU, the diagnosis of acute CVST is more likely. Applying both numbers to our sample, 1.3% (17 cases) had an HU greater than 62, where only a single case had an HU greater than 70 and was found to be a case of meningitis as we are discussing later on in this article. Comparing our results to those of both Black et al4 and Buyck et al,2 our study showed that using HU values to assess possibility of venous sinus clot on NCCT can be associated with less number of falsepositive readings than those resulting if only H:H ratio
N.T. AL-RYALAT ET AL.
is used. Our data also uphold using an HU greater than 70 as a strong predictor of CVST, as only one of our control populations had an average HU greater than 70. One of the main concerns of our study was to assess the effect of age as an independent factor on attenuation of cerebral venous sinuses, a correlation that is not established in the literature yet. Our results showed a decrease in HCT and Hb levels with age; this contribution can be explained by a recent study13 which found that aging is responsible for a gradual loss of kidney function that may lead to a decrease in Hb and HCT levels. Having demonstrated the effect of HCT on attenuation, this age-related decrease in HCT may underestimate the actual attenuation level of cerebral venous sinuses. Controlling HCT variant to eliminate its effect, we found that average attenuation is significantly and positively correlated with patient’s age; thus, the baseline attenuation of venous sinuses on NCCT of an older patient is higher than that for a young patient, considering that both have similar HCT level. The mechanism behind this can, to a certain extent, be explained by Carallo et al’s study,14 which concluded that blood viscosity increases with age, and this increase is disjoined from the contribution of HCT and plasma. Our finding can be applied clinically as elderly patients with clinical presentation suspicious for CVST are more likely to have normally appearing venous sinuses on NCCT, as the age-related increase in baseline attenuation can mask the thrombus; therefore, further confirmatory studies may still be warranted. Nonetheless, patient’s age can help determine whether venous sinus density is within normal ranges of the corresponding age or it is pathologically increased. This draws the attention of radiologists to the necessity of considering patient’s age when correlating laboratory results with the attenuation value and giving the final decision of confirming or excluding diagnosis of CVST. Our finding is also supported by Black et al’s study,4 which included 8 patients diagnosed with CVST. In their study, patient #2 (26 years old) and patient #7 (50 years old) have almost the same HCT value (36.2%, 36.5%) with different HU readings (86, 95) respectively. Applying our results to their patients, the difference in HU values between the 2 patients despite having the same HCT can now be partly attributed to the age-related nonthrombotic increase in HU value for the elderly patient. Further research is recommended in this regard to fully understand the exact mechanism for the effect of age on the attenuation of blood on NCCT. Elevated BUN/Cr ratio is commonly used as an indicator for dehydration. It was extensively used by many studies in their inclusion–exclusion criteria to detect dehydrated patients among their sample populations.2,4,15 Having a large set of data and knowing that many causes, other than dehydration, can increase BUN/Cr ratio (e.g., high-protein diet, steroid therapy, parenteral nutrition, gastrointestinal bleeding16), we tried to assess the feasibility
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
of using a high BUN/Cr ratio as a decisive parameter to exclude dehydrated patients when carrying out any study on assessment of attenuation on brain NCCT. In our study, we excluded patients who were clinically dehydrated regardless of their BUN/Cr ratio levels, and then we assessed the correlation between BUN/Cr ratio and average attenuation of venous sinuses. The association between the 2 parameters was insignificant. Despite being labeled as the most sensitive laboratory test for diagnosis of dehydration,17 the accuracy of BUN/Cr ratio is still beyond that of clinical assessment of the hydration status, as large number of patients might be missed if the diagnosis was exclusively based on BUN/Cr ratio level. Accordingly, we can conclude that BUN/Cr ratio should not be used as the only indicator of dehydration when carrying out a study on assessment of attenuation of vasculature on brain NCCT. Our results also showed a significant correlation between gender discrepancy and average attenuation of venous sinuses on NCCT, as the mean density for men was
5
significantly higher than that for women (53.7 HU and 52 HU, respectively). It is reported in the literature that CVST is much more common in women than men with a ratio of 3:1, and that this female preponderance is attributed to gender-specific risk factors such as pregnancy and puerperium.18 Despite this female predilection, our results showed that in the normal setting, men have a significantly higher baseline of cerebral venous sinus attenuation on NCCT than women (Table 1). To sum up, it is now clear why an apparent difference in attenuation can be observed upon comparing an NCCT of an old polycythemic man with that of a young anemic woman as demonstrated in Figure 1, which shows how wide the range of normal attenuation can be if the aforementioned factors were taken into account. Interestingly, one of our patients, a 48-year-old woman, who presented with headache, dizziness, visual disturbance, focal neurologic deficits, and impaired level of consciousness underwent the initial workup to rule out CVST. Her brain NCCT was free of masses but showed
Figure 1. (A) NCCT axial section of an elderly polycythemic man showing superior sagittal sinus with a mean density of 58.2 HU. (B) NCCT axial section of an elderly polycythemic man showing right and left sigmoid sinuses with a mean density of 67.75 HU and 50 HU, respectively. (C) NCCT axial section of a young anemic woman showing superior sagittal sinus with a mean density of 49 HU. (D) NCCT axial section of a young anemic woman showing right and left sigmoid sinuses with a mean density of 33 HU and 45 HU, respectively. Abbreviations: HU, Hounsfield units; NCCT, noncontrasted computed tomography.
ARTICLE IN PRESS N.T. AL-RYALAT ET AL.
6
Figure 2. (A) NCCT axial section for patient with meningitis showing hyperdense superior sagittal sinus with an extended falx cerebri thickening. (B) NCCT axial section for patient with cerebrovenous sinus thrombosis showing hyperdense superior sagittal sinus. Abbreviation: NCCT, noncontrasted computed tomography.
increased attenuation in the superior sagittal sinus, right sigmoid sinus, left sigmoid sinus, and vein of Galen with HU values of 72.0, 71.0, 81.8, and 70.2, respectively. Laboratory results showed normal BUN/Cr ratio, Hb, and HCT levels. Confirmatory MRV excluded any possibility of CVST, and based on clinical and laboratory measures, diagnosis of bacterial meningitis was made. We noticed that the HU values of our patient were all above 70, which is the value suggested by Black et al4 to be the cutoff between CVST and normal patients. Given the rarity of CVST, we suggest that meningitis should be added to the deferential diagnosis list of patients who present with new-onset neurological symptoms and have hyperattenuated cerebral venous sinuses on NCCT. Figure 2 shows the difference in the appearance of hyperattenuated cerebral venous sinuses on NCCT between CVST and meningitis patients. Dural thickening is observed in meningitis patients but not in CVST patients. Worth mentioning, the potential for generalizability of our findings to be applied to different populations and to include blood of different vascular structures arises from 2 facts. First, the mean HU value of our sample population is comparable to those of the control samples of 2 recent studies that were done on 2 different European populations.2,8 Second, NCCT has long been assumed to be a very useful modality to diagnose anemia by measuring the HU values of different vascular structures in the body19-21; this is the function of HCT factor being able to affect the attenuation of the whole blood in the body in a similar fashion to its effect on attenuation of cerebral venous sinuses that was appreciated in many other studies2,4,11,12 as well as in our study. We recommend future studies to assess the possibility for generalizing the effect of the aforementioned factors to include attenuation of vasculature of the whole body in a similar way to their effect on cerebral venous sinuses as we demonstrated in this manuscript.
Our study is hampered by some limitations that need to be taken into account. The lack of sufficient prior research on the same topic, especially the effect of age and BUN/Cr ratio on dural sinus attenuation, has made the role of this study to be predominantly exploratory regarding both factors. This also hampered the assessment of generalizability of our findings to vasculature of the whole body. We believe that having a preliminary background to compare with our findings could have widened the interpretive role of our study. Thus, this manuscript offers a groundwork for future research to build on in order to reach a good understanding of all factors that can affect attenuation of blood on brain as well as wholebody NCCT scans.
Conclusion Absolute measurement of dural sinus attenuation provides more reliable findings and better utilization of brain NCCT. Correlating all possible patient-related factors like HCT, patient’s age, and gender upon interpreting brain NCCT increases the confidence of the radiologists and minimizes the number of unnecessary confirmatory investigations, especially in emergency settings and in situations where NCCT is the only available diagnostic modality. HCT factor has long been known to affect attenuation, a finding that is also appreciated in our study. Our findings also demonstrate that dural sinus attenuation correlates with patient’s age, an effect that should be taken into account upon interpreting any brain NCCT. Patient’s age can also help determine whether dural sinus attenuation is within normal ranges or it is pathologically increased. We also suggest that BUN/Cr ratio should not be used as the only indicator of dehydration when carrying out a study on assessment of dural sinus attenuation on brain NCCT.
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
References 1. Guenther G, Arauz A. Cerebral venous thrombosis: a diagnostic and treatment update. Neurologia 2011;26:488498. 2. Buyck PJ, De Keyzer F, Vanneste D, et al. CT density measurement and H:H ratio are useful in diagnosing acute cerebral venous sinus thrombosis. AJNR Am J Neuroradiol 2013;34:1568-1572. 3. Saposnik G, Barinagarrementeria F, Brown RD Jr, et al. Diagnosis and management of cerebral venous thrombosis: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42:1158-1192. 4. Black DF, Rad AE, Gray LA, et al. Cerebral venous sinus density on noncontrast CT correlates with hematocrit. AJNR Am J Neuroradiol 2011;32:1354-1357. 5. Goldstein M, Quen L, Jacks L, et al. Acute abdominal venous thromboses—the hyperdense CT sign. J Comput Assist Tomogr 2012;36:8-13. 6. Hsu CT, Wang ZJ, Yu AS, et al. Physiology of renal medullary tip hyperattenuation at unenhanced CT: urinary specific gravity and the NaCl concentration gradient. Radiology 2008;247:147-153. 7. Roland T, Jacobs J, Rappaport A, et al. Unenhanced brain CT is useful to decide on further imaging in suspected venous sinus thrombosis. Clin Radiol 2010;65:34-39. 8. Alsafi A, Lakhani A, Carlton Jones L, et al. Cerebral venous sinus thrombosis, a nonenhanced CT diagnosis? Radiol Res Pract 2015;2015:581437. 9. Provenzale JM, Kranz PG. Dural sinus thrombosis: sources of error in image interpretation. AJR Am J Roentgenol 2011;196:23-31. 10. Stein J, Huerta K. When looking at a non-contrast head CT, what actually appears white in an acute hemorrhagic stroke? Cal J Emerg Med 2002;3:70-71.
7 11. Healy JF, Nichols C. Polycythemia mimicking venous sinus thrombosis. AJNR Am J Neuroradiol 2002;23:14021403. 12. Bruni SG, Patafio FM, Dufton JA, et al. The assessment of anemia from attenuation values of cranial venous drainage on unenhanced computed tomography of the head. Can Assoc Radiol J 2013;64:46-50. 13. Weinstein JR, Anderson S. The aging kidney: physiological changes. Adv Chronic Kidney Dis 2010;17:302-307. 14. Carallo C, Irace C, De Franceschi MS, et al. The effect of aging on blood and plasma viscosity. An 11.6 years follow-up study. Clin Hemorheol Microcirc 2011;47:6774. 15. Liu CH, Lin SC, Lin JR, et al. Dehydration is an independent predictor of discharge outcome and admission cost in acute ischaemic stroke. Eur J Neurol 2014;21:1184-1191. 16. Uchino S, Bellomo R, Goldsmith D. The meaning of the blood urea nitrogen/creatinine ratio in acute kidney injury. Clin Kidney J 2012;5:187-191. 17. Riccardi A, Chiarbonello B, Minuto P, et al. Identification of the hydration state in emergency patients: correlation between caval index and BUN/creatinine ratio. Eur Rev Med Pharmacol Sci 2013;17:1800-1803. 18. Ozcan TA, Meral H, Ozben S, et al. Cerebral venous sinus thrombosis: gender differences in ten years experience. J Psychiatry Neurol Sci 2013;26:281-285. 19. Kamel EM, Rizzo E, Duchosal MA, et al. Radiological profile of anemia on unenhanced MDCT of the thorax. Eur Radiol 2008;18:1863-1868. 20. Jung C, Groth M, Bley TA, et al. Assessment of anemia during CT pulmonary angiography. Eur J Radiol 2012; 81:4196-4202. 21. Lan H, Nishihara S, Nishitani H. Accuracy of computed tomography attenuation measurements for diagnosing anemia. Jpn J Radiol 2010;28:53-57.
Factors Affecting Attenuation of Dural Sinuses on Noncontrasted Computed Tomography Scan Nosaiba T. Al-Ryalat, MD (Doctor of Medicine),* Saif Aldeen S. AlRyalat, MS (Medical Student),† Lna W. Malkawi, MS,† Ethar F. Al-Zeena, MD,† Mahasen S. Al Najar, MD,* and Azmy M. Hadidy, MD, FCRC (Fellow of the Royal Collage of Radiologists)*
Background and Purpose: Noncontrasted computed tomography (NCCT) is used as the initial neuroimaging test of choice for patients who present with newonset neurological symptoms. An apparently hyperattenuated venous sinus may lead to the suspicion of cerebral venous sinus thrombosis (CVST). Improved understanding of all factors that can affect attenuation of dural sinuses can guide triage of patients to or from further investigations of suspected CVST. The purpose of this retrospective study was to assess the effect of different factors including hematocrit (HCT), hemoglobin (Hb), age, BUN/Cr ratio (blood urea nitrogen-tocreatinine ratio), and gender on the attenuation of dural sinuses on brain NCCT. Methods: A total of 1293 patients with neurological symptoms who presented to the emergency department were included in this study. For each patient, clinical assessment, laboratory investigations, and brain NCCT were reviewed. For each brain NCCT, the average attenuation of superior sagittal sinus and both right and left sigmoid sinuses was measured. Results: Positive significant correlations were found between average attenuation of dural sinuses on one hand and each of age, Hb, and HCT on the other hand. No significant correlation was found between average attenuation and BUN/Cr ratio. Gender discrepancy was also significant as higher attenuation was found in men. Conclusion: Age, gender, and Hb levels are the main factors that should be taken into account upon the assessment of dural sinuses on brain NCCT. The highest normal attenuation is predicted in an elderly polycythemic man and the lowest is predicted in a young anemic woman. Key Words: Cerebrovenous sinus thrombosis—noncontrasted CT—dural sinuses—attenuation—age. © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved.
From the *Radiology and Nuclear Medicine Department, Jordan University Hospital, The University of Jordan, Amman, Jordan; and †Faculty of Medicine, Jordan University Hospital, The University of Jordan, Amman, Jordan. Received March 24, 2016; revision received May 18, 2016; accepted July 2, 2016. Address correspondence to Saif Aldeen S. AlRyalat, MS (Medical Student), Faculty of Medicine, Jordan University Hospital, The University of Jordan, P.O. Box 1669 Tela Al Ali, 11953 Amman, Jordan. E-mail: [email protected]. 1052-3057/$ - see front matter © 2016 National Stroke Association. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2016.07.002
Introduction Cerebral venous sinus thrombosis (CVST) is a challenging neurologic disorder that represents approximately 0.5% of all stroke cases worldwide.1 The symptoms and clinical features are nonspecific leading to a highly variable presentation among different age groups, making the diagnosis of CVST mainly based on a high index of clinical suspicion and imaging confirmation. Despite being a rare neurologic disorder, with an annual incidence estimated at 2-7 cases per 1 million populations,2 mortality rates of 10% have been reported among CVST patients,1 making early diagnosis and treatment very critical.
Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2016: pp ■■–■■
1
ARTICLE IN PRESS N.T. AL-RYALAT ET AL.
2
Noncontrasted computed tomography (NCCT) scan is widely used as the initial neuroimaging test of choice for patients who present to the emergency department with new-onset neurological symptoms such as headache, seizures, mental alteration, or focal neurological signs.3 As subjective assessment of the radiodensity based on apparent increase or decrease in attenuation may be misleading,2 radiologists tend routinely to measure the attenuation of intracerebral structures on brain NCCT by means of Hounsfield units (HU), where observing a dural sinus with an elevated HU value (i.e., hyperattenuation) may lead to the suspicion of CVST.4 These hyperattenuation signs have a crucial role in detecting CVST at an early stage when treatment is most likely to be effective and to greatly affect the clinical outcome.5 The drawback of these signs is that many factors can affect the degree of attenuation of blood, therefore confusing normality with pathologic conditions and thus increasing the number of false-positive and false-negative readings upon interpretation of NCCT. Improved understanding of all factors that can affect attenuation of dural sinuses can guide triage of patients to or from further investigations of suspected CVST. Despite the crucial role of NCCT, still minimal information is known about these factors. Hemoconcentration and dehydration have long been assumed as known factors to increase the attenuation of blood on NCCT,2,4,6 but the correlation with many other factors is not established yet. The purpose of this retrospective study, which included one of the largest samples of control population ever used for the same purpose, is to assess the effect of different factors including hematocrit (HCT), hemoglobin (Hb), age, BUN/Cr ratio (blood urea nitrogen-to-creatinine ratio), and gender on the attenuation of dural sinuses on brain NCCT.
Methodology This retrospective study was approved by our institutional review board; patient informed consent was waived.
Patients A total of 1580 patients who presented to the emergency department in our tertiary hospital in the period between June 28, 2014, and September 5, 2015, were recruited in this study; clinical assessment, laboratory investigations, and NCCT of brain were reviewed for each patient. To meet our inclusion criteria, (1) each patient was presented with new-onset neurological symptoms such as headache, seizure, mental alteration, or focal neurological signs; (2) NCCT scan, complete blood count (CBC), and kidney function test (KFT) were done within 24 hours of onset of symptoms; and (3) diagnosis of CVST was excluded by magnetic resonance venography (MRV), or diagnosis other than CVST was established. The following were the exclusion criteria: (1) clinically dehydrated patients as indicated in their hydration
assessment sheet; (2) hemorrhage or skull fracture adjacent to dural sinuses, severe brain swelling, intra-axial or extra-axial mass, contrast media administration within the previous 24 hours, or patients who received blood transfusion within the previous 24 hours; and (3) neonates below 6 months in order to eliminate the effect of fetal Hb. Up to 1293 patients were included in our study. Of these patients, 286 were excluded according to our criteria: 142 patients due to clinical dehydration; 93 due to head pathologies mentioned in criteria #2; and 51 due to age less than 6 months, with only 1 excluded due to CVST confirmed by MRV. Male–female difference was not significant as 671 (51.9%) were male and 622 (48.1%) were female. Of the 1293 patients included, 413 presented with various neurological signs and symptoms including headache, dizziness, nausea, and vomiting without focal neurological signs; they had normal CT and normal magnetic resonance venography later on (so CVST was excluded). Two hundred five patients had focal neurological signs and negative CT, had their MRI done, and showed an infarction; furthermore, other 162 patients had focal neurological signs and negative CT and MRI and diagnosed with transient ischemic attack. Up to 164 patients presented mainly with headache and diagnosed with sinusitis on CT. One hundred seventeen patients presented with neurological signs and symptoms with fever and negative CT and diagnosed with lumber puncture with meningitis. One hundred nine patients presented with headache and high blood pressure, had their brain CT done, and showed no brain hemorrhage. One hundred twenty-three patients presented with neurological symptoms and found to have other system-related infections.
Data Interpretation Attenuation of superior sagittal sinus and both left and right sigmoid sinuses was measured by 1 specialized radiologist in order to eliminate interobserver variation, 1 region of interest (ROI) measuring 2-mm recorder used to measure attenuation of each sinus in HU, except in small nondominant sigmoid sinuses in which we used an ROI of 1 mm. The average sinus density was calculated by taking the mean of density of superior sagittal sinus and both left and right sigmoid sinuses. Hb and HCT were obtained from CBC whereas Cr and BUN were obtained from the KFT. BUN/Cr ratio was calculated. The Hounsfield number of superior sagittal sinus to HCT level (H:H ratio, Hounsfield unit-tohematocrit ratio) was calculated.
Imaging Protocol All images were done via Siemens, Somatom Definition Flash 128 slice, CT scanner (Siemens Medical Solutions, Forchheim, Germany), with the following parameters: 360-410 mAs, 120 kV, section thickness of 5 mm, and reconstruction increment of 5 mm.
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
Statistical Analysis In our statistical analysis, we used IBM SPSS for Windows release 21.0 (SPSS Inc., Chicago, IL); descriptive statistics was used to describe our sample population. Correlation test was used to study the relation between age and HCT; partial correlation tests were used to study the relation between average attenuation and each of the following: age, HCT, Hb, and BUN/Cr ratio (by controlling other factors when studying each one); independentsample t-test was used to study the relation between average attenuation and gender, it was also used to compare BUN/Cr ratios above and below 20. Multivariate and univariate analyses were used to compare different variables including age, HCT, average attenuation, and BUN/Cr ratio with gender. A P value < .05 was taken as significant.
Results A total of 1293 patients from all ages were included with a mean age of 55.5 years (ranged from 1 to 97 years) without significant gender difference (mean age for men was 55.1 years and for women was 55.8 years).
HCT/Hb The mean HCT for our sample population was 39.4 (ranged from 13.6 to 58.9) whereas for Hb the mean was 12.9 (ranged from 6.3 to 18.7). Negative significant correlation (−.12) was found between age and HCT with a P value < .001; this correlation was also found to become weaker (−.07) but significant (P = .008) upon controlling Cr. Positive significant correlation (.48) was found between the average attenuation and HCT with a P value < .001; a stronger correlation was found between average attenuation and Hb (.50) with a P value < .001. Mean values of attenuation for anemic, normal, and polycythemic patients are mentioned in Table 1. To calculate the H:H ratio, we used the attenuation values of superior sagittal sinus in order to compare our ratio Table 1. (a) Male and (b) female densities HCT/age (a) Anemic Normal Polycythemic (b) Anemic Normal Polycythemic
<18
18-65
>65
49.6 53.5 55.4
55.0 55.5 56.2
52.5 55.0 61.4
47.9 50.9 53.7
50.7 53.2 55.6
51.8 52.9 59.7
Abbreviation: HCT, hematocrit. (a) Anemic: HCT < 40, normal: 40 < HCT < 54, polycythemic: HCT > 54. (b) Anemic: HCT < 36, normal: 36 < HCT < 46, polycythemic: HCT > 46.
3
with those in other studies; the maximum H:H ratio reached 3.82, where 31 patients had a ratio of more than 2.
Age Positive and significant correlation (.12) was found between age and average attenuation with a P value < .001; this correlation was found upon controlling the effect of HCT. The mean attenuation for different age groups for both men and women is shown in Table 1, a,b.
BUN/Cr Ratio The mean BUN/Cr ratio for our sample was 16.3. Correlation between average attenuation on one hand and each of BUN/Cr ratio, BUN, and Cr on the other hand was found to be insignificant with P values of .27, .55, and .43, respectively. No significant difference was found between attenuation of patients with BUN/Cr ratio above 20 and of those with a ratio below 20 (P value = .11).
Gender On a multivariate test, a significant P value < .001 was found. On Levene’s test of equality, the significance was only for HCT (P < .001) and average attenuation (P = .01); no significance was found for age or BUN/Cr ratio. Gender discrepancy was found to be significant regarding average attenuation using partial correlation test (with controlling the effect of HCT) with a P value = .02. The mean attenuation for men was 53.9 HU and for women was 52.4 HU (Table 1).
Discussion NCCT scan is performed as the first-line imaging modality in the emergency setting of most institutions for patients who present with new-onset neurological symptoms; this is the function of NCCT being a cost-effective, fast, and widely available imaging technique.2,7 The classic sign of acute CVST on NCCT is an increase in attenuation of the thrombosed venous sinus.2 Because a moderate increase in attenuation may not be always picked up by the radiologists, measurement of attenuation values of the sinuses by means of HU might be a more objective and reliable method. A recent study estimated the sensitivity of NCCT in the detection of CVST at 73%7; this may be even lower in daily clinical practice where multiple patientrelated factors can affect the attenuation of cerebral vasculature increasing possibility of false-positive or falsenegative interpretation of sinus attenuation. Where a prompt diagnosis of CVST is of critical importance as anticoagulation therapy is thought to prevent propagation of the thrombus,8 a false-positive diagnosis can, on the other hand, cause a delay in establishing the correct diagnosis as well as proceeding to inappropriate treatment.9 Using a large sample of controls with a widely variable range of attenuation values, we tried to assess
ARTICLE IN PRESS 4
the effect of many factors on attenuation of venous sinuses trying to reach the maximum possible sensitivity for NCCT in the diagnosis of CVST. Starting with HCT and Hb levels, a significant association between blood attenuation and both Hb and HCT levels has long been established.2,4,10 This correlation is also appreciated in our study which showed a positive and significant correlation between average density of dural sinuses and both Hb and HCT levels; a stronger linear relationship with Hb than that with HCT was demonstrated. It was reported that patients with polycythemia vera, who have high Hb and HCT levels, have hyperattenuated blood.11 Furthermore, blood of anemic patients, who have low Hb levels, was found to appear hypoattenuated on NCCT.12 Our study included both anemic and polycythemic patients in addition to patients with normal Hb; Table 1, a,b shows the mean sinus attenuation for each group. Black et al4 suggested the potential usefulness of HU values and H:H ratio in the diagnosis of CVST; they found that hemoconcentration correlates with attenuation of cerebral venous sinuses on NCCT. They were the first to use the H:H ratio to assess the correlation between hyperattenuation on NCCT and HCT level and recommended an H:H ratio greater than or equal to 2 as a cutoff value to separate patients with CVST from those without. Applying the H:H threshold suggested by Black et al to our study, which included a large sample of controls without CVST, 31 cases (2.5%) had an H:H ratio greater than or equal to 2 (ranging from 2.00 to 3.67), a large number that might be easily misinterpreted as cases of CVST increasing number of false-positive readings. A subsequent study suggested an H:H ratio greater than 1.52 to indicate a strong likelihood of a clot,2 a lower threshold that will result in a higher number of false-positive readings if HCT was the only factor taken into consideration upon interpreting a brain NCCT. Although they successfully controlled the effect of HCT factor by using the H:H ratio, other factors are still affecting the attenuation level giving rise to this large number of potentially false-positive readings; these factors also need to be corrected before reaching the final diagnosis. Noticeably, Black et al suggested an HU greater than 70 to be suspicious for CVST and to warrant further confirmatory investigations. Followed by Buyck et al,2 a more recent study suggested that in patients with an attenuation of more than 62 HU, the diagnosis of acute CVST is more likely. Applying both numbers to our sample, 1.3% (17 cases) had an HU greater than 62, where only a single case had an HU greater than 70 and was found to be a case of meningitis as we are discussing later on in this article. Comparing our results to those of both Black et al4 and Buyck et al,2 our study showed that using HU values to assess possibility of venous sinus clot on NCCT can be associated with less number of falsepositive readings than those resulting if only H:H ratio
N.T. AL-RYALAT ET AL.
is used. Our data also uphold using an HU greater than 70 as a strong predictor of CVST, as only one of our control populations had an average HU greater than 70. One of the main concerns of our study was to assess the effect of age as an independent factor on attenuation of cerebral venous sinuses, a correlation that is not established in the literature yet. Our results showed a decrease in HCT and Hb levels with age; this contribution can be explained by a recent study13 which found that aging is responsible for a gradual loss of kidney function that may lead to a decrease in Hb and HCT levels. Having demonstrated the effect of HCT on attenuation, this age-related decrease in HCT may underestimate the actual attenuation level of cerebral venous sinuses. Controlling HCT variant to eliminate its effect, we found that average attenuation is significantly and positively correlated with patient’s age; thus, the baseline attenuation of venous sinuses on NCCT of an older patient is higher than that for a young patient, considering that both have similar HCT level. The mechanism behind this can, to a certain extent, be explained by Carallo et al’s study,14 which concluded that blood viscosity increases with age, and this increase is disjoined from the contribution of HCT and plasma. Our finding can be applied clinically as elderly patients with clinical presentation suspicious for CVST are more likely to have normally appearing venous sinuses on NCCT, as the age-related increase in baseline attenuation can mask the thrombus; therefore, further confirmatory studies may still be warranted. Nonetheless, patient’s age can help determine whether venous sinus density is within normal ranges of the corresponding age or it is pathologically increased. This draws the attention of radiologists to the necessity of considering patient’s age when correlating laboratory results with the attenuation value and giving the final decision of confirming or excluding diagnosis of CVST. Our finding is also supported by Black et al’s study,4 which included 8 patients diagnosed with CVST. In their study, patient #2 (26 years old) and patient #7 (50 years old) have almost the same HCT value (36.2%, 36.5%) with different HU readings (86, 95) respectively. Applying our results to their patients, the difference in HU values between the 2 patients despite having the same HCT can now be partly attributed to the age-related nonthrombotic increase in HU value for the elderly patient. Further research is recommended in this regard to fully understand the exact mechanism for the effect of age on the attenuation of blood on NCCT. Elevated BUN/Cr ratio is commonly used as an indicator for dehydration. It was extensively used by many studies in their inclusion–exclusion criteria to detect dehydrated patients among their sample populations.2,4,15 Having a large set of data and knowing that many causes, other than dehydration, can increase BUN/Cr ratio (e.g., high-protein diet, steroid therapy, parenteral nutrition, gastrointestinal bleeding16), we tried to assess the feasibility
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
of using a high BUN/Cr ratio as a decisive parameter to exclude dehydrated patients when carrying out any study on assessment of attenuation on brain NCCT. In our study, we excluded patients who were clinically dehydrated regardless of their BUN/Cr ratio levels, and then we assessed the correlation between BUN/Cr ratio and average attenuation of venous sinuses. The association between the 2 parameters was insignificant. Despite being labeled as the most sensitive laboratory test for diagnosis of dehydration,17 the accuracy of BUN/Cr ratio is still beyond that of clinical assessment of the hydration status, as large number of patients might be missed if the diagnosis was exclusively based on BUN/Cr ratio level. Accordingly, we can conclude that BUN/Cr ratio should not be used as the only indicator of dehydration when carrying out a study on assessment of attenuation of vasculature on brain NCCT. Our results also showed a significant correlation between gender discrepancy and average attenuation of venous sinuses on NCCT, as the mean density for men was
5
significantly higher than that for women (53.7 HU and 52 HU, respectively). It is reported in the literature that CVST is much more common in women than men with a ratio of 3:1, and that this female preponderance is attributed to gender-specific risk factors such as pregnancy and puerperium.18 Despite this female predilection, our results showed that in the normal setting, men have a significantly higher baseline of cerebral venous sinus attenuation on NCCT than women (Table 1). To sum up, it is now clear why an apparent difference in attenuation can be observed upon comparing an NCCT of an old polycythemic man with that of a young anemic woman as demonstrated in Figure 1, which shows how wide the range of normal attenuation can be if the aforementioned factors were taken into account. Interestingly, one of our patients, a 48-year-old woman, who presented with headache, dizziness, visual disturbance, focal neurologic deficits, and impaired level of consciousness underwent the initial workup to rule out CVST. Her brain NCCT was free of masses but showed
Figure 1. (A) NCCT axial section of an elderly polycythemic man showing superior sagittal sinus with a mean density of 58.2 HU. (B) NCCT axial section of an elderly polycythemic man showing right and left sigmoid sinuses with a mean density of 67.75 HU and 50 HU, respectively. (C) NCCT axial section of a young anemic woman showing superior sagittal sinus with a mean density of 49 HU. (D) NCCT axial section of a young anemic woman showing right and left sigmoid sinuses with a mean density of 33 HU and 45 HU, respectively. Abbreviations: HU, Hounsfield units; NCCT, noncontrasted computed tomography.
ARTICLE IN PRESS N.T. AL-RYALAT ET AL.
6
Figure 2. (A) NCCT axial section for patient with meningitis showing hyperdense superior sagittal sinus with an extended falx cerebri thickening. (B) NCCT axial section for patient with cerebrovenous sinus thrombosis showing hyperdense superior sagittal sinus. Abbreviation: NCCT, noncontrasted computed tomography.
increased attenuation in the superior sagittal sinus, right sigmoid sinus, left sigmoid sinus, and vein of Galen with HU values of 72.0, 71.0, 81.8, and 70.2, respectively. Laboratory results showed normal BUN/Cr ratio, Hb, and HCT levels. Confirmatory MRV excluded any possibility of CVST, and based on clinical and laboratory measures, diagnosis of bacterial meningitis was made. We noticed that the HU values of our patient were all above 70, which is the value suggested by Black et al4 to be the cutoff between CVST and normal patients. Given the rarity of CVST, we suggest that meningitis should be added to the deferential diagnosis list of patients who present with new-onset neurological symptoms and have hyperattenuated cerebral venous sinuses on NCCT. Figure 2 shows the difference in the appearance of hyperattenuated cerebral venous sinuses on NCCT between CVST and meningitis patients. Dural thickening is observed in meningitis patients but not in CVST patients. Worth mentioning, the potential for generalizability of our findings to be applied to different populations and to include blood of different vascular structures arises from 2 facts. First, the mean HU value of our sample population is comparable to those of the control samples of 2 recent studies that were done on 2 different European populations.2,8 Second, NCCT has long been assumed to be a very useful modality to diagnose anemia by measuring the HU values of different vascular structures in the body19-21; this is the function of HCT factor being able to affect the attenuation of the whole blood in the body in a similar fashion to its effect on attenuation of cerebral venous sinuses that was appreciated in many other studies2,4,11,12 as well as in our study. We recommend future studies to assess the possibility for generalizing the effect of the aforementioned factors to include attenuation of vasculature of the whole body in a similar way to their effect on cerebral venous sinuses as we demonstrated in this manuscript.
Our study is hampered by some limitations that need to be taken into account. The lack of sufficient prior research on the same topic, especially the effect of age and BUN/Cr ratio on dural sinus attenuation, has made the role of this study to be predominantly exploratory regarding both factors. This also hampered the assessment of generalizability of our findings to vasculature of the whole body. We believe that having a preliminary background to compare with our findings could have widened the interpretive role of our study. Thus, this manuscript offers a groundwork for future research to build on in order to reach a good understanding of all factors that can affect attenuation of blood on brain as well as wholebody NCCT scans.
Conclusion Absolute measurement of dural sinus attenuation provides more reliable findings and better utilization of brain NCCT. Correlating all possible patient-related factors like HCT, patient’s age, and gender upon interpreting brain NCCT increases the confidence of the radiologists and minimizes the number of unnecessary confirmatory investigations, especially in emergency settings and in situations where NCCT is the only available diagnostic modality. HCT factor has long been known to affect attenuation, a finding that is also appreciated in our study. Our findings also demonstrate that dural sinus attenuation correlates with patient’s age, an effect that should be taken into account upon interpreting any brain NCCT. Patient’s age can also help determine whether dural sinus attenuation is within normal ranges or it is pathologically increased. We also suggest that BUN/Cr ratio should not be used as the only indicator of dehydration when carrying out a study on assessment of dural sinus attenuation on brain NCCT.
ARTICLE IN PRESS FACTORS AFFECTING DURAL SINUS ATTENUATION
References 1. Guenther G, Arauz A. Cerebral venous thrombosis: a diagnostic and treatment update. Neurologia 2011;26:488498. 2. Buyck PJ, De Keyzer F, Vanneste D, et al. CT density measurement and H:H ratio are useful in diagnosing acute cerebral venous sinus thrombosis. AJNR Am J Neuroradiol 2013;34:1568-1572. 3. Saposnik G, Barinagarrementeria F, Brown RD Jr, et al. Diagnosis and management of cerebral venous thrombosis: a statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42:1158-1192. 4. Black DF, Rad AE, Gray LA, et al. Cerebral venous sinus density on noncontrast CT correlates with hematocrit. AJNR Am J Neuroradiol 2011;32:1354-1357. 5. Goldstein M, Quen L, Jacks L, et al. Acute abdominal venous thromboses—the hyperdense CT sign. J Comput Assist Tomogr 2012;36:8-13. 6. Hsu CT, Wang ZJ, Yu AS, et al. Physiology of renal medullary tip hyperattenuation at unenhanced CT: urinary specific gravity and the NaCl concentration gradient. Radiology 2008;247:147-153. 7. Roland T, Jacobs J, Rappaport A, et al. Unenhanced brain CT is useful to decide on further imaging in suspected venous sinus thrombosis. Clin Radiol 2010;65:34-39. 8. Alsafi A, Lakhani A, Carlton Jones L, et al. Cerebral venous sinus thrombosis, a nonenhanced CT diagnosis? Radiol Res Pract 2015;2015:581437. 9. Provenzale JM, Kranz PG. Dural sinus thrombosis: sources of error in image interpretation. AJR Am J Roentgenol 2011;196:23-31. 10. Stein J, Huerta K. When looking at a non-contrast head CT, what actually appears white in an acute hemorrhagic stroke? Cal J Emerg Med 2002;3:70-71.
7 11. Healy JF, Nichols C. Polycythemia mimicking venous sinus thrombosis. AJNR Am J Neuroradiol 2002;23:14021403. 12. Bruni SG, Patafio FM, Dufton JA, et al. The assessment of anemia from attenuation values of cranial venous drainage on unenhanced computed tomography of the head. Can Assoc Radiol J 2013;64:46-50. 13. Weinstein JR, Anderson S. The aging kidney: physiological changes. Adv Chronic Kidney Dis 2010;17:302-307. 14. Carallo C, Irace C, De Franceschi MS, et al. The effect of aging on blood and plasma viscosity. An 11.6 years follow-up study. Clin Hemorheol Microcirc 2011;47:6774. 15. Liu CH, Lin SC, Lin JR, et al. Dehydration is an independent predictor of discharge outcome and admission cost in acute ischaemic stroke. Eur J Neurol 2014;21:1184-1191. 16. Uchino S, Bellomo R, Goldsmith D. The meaning of the blood urea nitrogen/creatinine ratio in acute kidney injury. Clin Kidney J 2012;5:187-191. 17. Riccardi A, Chiarbonello B, Minuto P, et al. Identification of the hydration state in emergency patients: correlation between caval index and BUN/creatinine ratio. Eur Rev Med Pharmacol Sci 2013;17:1800-1803. 18. Ozcan TA, Meral H, Ozben S, et al. Cerebral venous sinus thrombosis: gender differences in ten years experience. J Psychiatry Neurol Sci 2013;26:281-285. 19. Kamel EM, Rizzo E, Duchosal MA, et al. Radiological profile of anemia on unenhanced MDCT of the thorax. Eur Radiol 2008;18:1863-1868. 20. Jung C, Groth M, Bley TA, et al. Assessment of anemia during CT pulmonary angiography. Eur J Radiol 2012; 81:4196-4202. 21. Lan H, Nishihara S, Nishitani H. Accuracy of computed tomography attenuation measurements for diagnosing anemia. Jpn J Radiol 2010;28:53-57.