DOI: 10.33962/roneuro-2021-001 On thin ice. Does season influence the ris k of haemorrhagic stroke from brain vascular malformations? Ioan Alexandru Florian, Teodora Larisa Timiș, Lehel Beni, Larisa Serban, Ioan Stefan Florian, Adrian Bălașa, Ioana Berindan-Neagoe Romanian Neurosurgery (2021) XXXV (1): pp. 7-13 DOI: 10.33962/roneuro-2021-001 www.journals.lapub.co.uk/index.php/roneurosurgery On thin ice. Does season influence the risk of haemorrhagic stroke from brain vascular malformations? Ioan Alexandru Florian1,2, Teodora Larisa Timiș3, Lehel Beni1, Larisa Serban1, Ioan Stefan Florian1,2, Adrian Bălașa4, Ioana Berindan-Neagoe5 1 Clinic of Neurosurgery, Cluj County Emergency Clinical Hospital, Cluj-Napoca, ROMANIA 2 Department of Neurosurgery, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, ROMANIA 3 Department of Physiology, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, ROMANIA 4 Clinic of Neurosurgery, Tîrgu Mureș County Clinical Emergency Hospital and Department of Neurosurgery, Tîrgu Mureș University of Medicine, Pharmacy, Science and Technology, Tîrgu Mureș, ROMANIA 5 Research Center for Functional Genomics, Biomedicine, and Translational Medicine, Institute of Doctoral Studies, "Iuliu Hatieganu" University of Medicine and Pharmacy, Cluj-Napoca and Department of Experimental Pathology "Prof. Ion Chiricuta", The Oncology Institute, Cluj-Napoca, ROMANIA ABSTRACT Introduction. Brain vascular malformations (BVMs) are congenital lesions with evolutive properties that possess a considerable chance of causing intracranial hemorrhage. The most common types are arteriovenous malformations (AVMs), aberrant entanglements of deformed vessels that shunt blood from the arteries directly into the veins, and cavernous malformations (CMs), being mulberry-shaped sinusoid spaces filled with blood. The rate of hemorrhagic stroke varies between these two types of lesions, being the most common form of symptomatic presentation for AVMs, but a much rarer occurrence for CMs. The purpose of our pilot study was to test whether the incidence of intracranial hemorrhage from BVMs varies between seasons, as well as examining a possible causality for this event. Material and methods. We performed a retrospective analysis on the cases of ruptured BVMs of the brain operated by the senior surgeon in our department between January 2008 and December 2019. We then divided the patients according to type of lesion and gender, based on the month of the year when their pathologies caused hemorrhagic stroke. We performed Pearson’s chi square test to verify the Keywords arteriovenous malformation (avm), cavernous malformation (cm), rupture, haemorrhagic stroke, surgical removal, gender, season, meteorological conditions Corresponding author: Ioan Alexandru Florian Department of Neurosurgery, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania florian.ioan.alexandru@gmail.com Copyright and usage. This is an Open Access article, distributed under the terms of the Creative Commons Attribution Non–Commercial No Derivatives License (https://creativecommons .org/licenses/by-nc-nd/4.0/) which permits non- commercial re-use, distribution, and reproduction in any medium, provided the original work is unaltered and is properly cited. The written permission of the Romanian Society of Neurosurgery must be obtained for commercial re-use or in order to create a derivative work. ISSN online 2344-4959 © Romanian Society of Neurosurgery First published March 2021 by London Academic Publishing www.lapub.co.uk http://www.lapub.co.uk/ 8 Ioan Alexandru Florian, Teodora Larisa Timiș, Lehel Beni et al. relationship between season and rate of rupture of AVMs and CMs, individual month and rate of rupture, season and gender, and individual month and gender. Results. There were 87 ruptured vascular malformations, out of which 71 were AVMs and 16 were CMs. There were 51 males (40 AVMs, 11 CMs) and 36 females (31 AVMs, 5 CMs). The majority of hemorrhagic strokes occurred in the months of July (10 AVVMs, 2 CMs) and December (10 AVMs, 1 CM). We obtained a statistically significant correlation between the summer season and presentation with ruptured cavernous malformation, as well as the male sex and presentation with a ruptured AVM in December, whereas the female sex presented a correlation with ruptured AVMs in the month of March. We also obtained a correlation between the male sex and presenting with a ruptured vascular malformation of any kind in December, as well as the female gender and hemorrhagic stroke from any vascular malformation in the months January and August. Conclusion. Despite promising statistical results, the relatively low number of cases may not be applicable to a larger patient population. It seems probable that meteorological conditions, especially extreme temperatures, might act as an additional risk factor for hemorrhagic stroke from vascular malformations, however these findings should be corroborated with supplementary case series from other centers, or a large prospective trial. INTRODUCTION Brain vascular malformations (BMVs) represent congenital aberrations of the cerebral blood vessels, having evolutive properties and a variable tendency to cause hemorrhagic stroke across its subtypes. The most common BMVs are arteriovenous malformations (AVMs), which consist of an entanglement of abnormal vessels that shunt blood from arterial feeders directly into one or more draining veins [4, 5, 10]. Hemorrhagic stroke stands as their most common form of symptomatic presentation, AVMs being also the leading cause of spontaneous intracranial hemorrhage in the young population. Cavernous malformations (CMs), also referred to as cavernomas, have a lower prevalence than AVMs, are mulberry-shaped sinusoid caverns filled with blood in various stages of hemolysis and are generally asymptomatic upon discovery [1, 2, 9]. Regarding symptoms, epilepsy is described as the most frequent for CMs, hemorrhagic stroke being slightly less common. Spontaneous intracranial hemorrhage (ICH), or hemorrhagic stroke, is responsible for roughly 20% of all stroke types [15, 20]. Despite medical and surgical advancements, it remains a mortifying pathology with a mortality as high as 30%, as well as a strikingly elevated morbidity among survivors. The influence of seasonal change on the propensity of hemorrhagic stroke has been widely investigated across several areas of the world, although the results are inconsistent and continuously debated. The majority of such studies described a rise in the incidence of stroke throughout the colder months of the year and a decline in the warmer months [3, 11, 12, 22]. Since AVM and CM rupture depends on both acquired and environmental factors, it is reasonable to assume that meteorological factors may indeed contribute to this event. Nevertheless, our current comprehension of the individual environmental triggers is narrow at best. Aside from risk factors such as smoking, alcohol and drug consumption, and various comorbidities like arterial hypertension and diabetes mellitus, we believe it important to also tackle the added influence of seasonal variation in hemorrhagic stroke from these lesions. In the present study, we examine the seasonal rates of rupture from BVMs which were admitted in our hospital in a 12-year interval. MATERIALS AND METHODS We conducted a retrospective analysis on the cases of ruptured BVMs (AVMs and CMs) operated by the senior surgeon in our department (the third author) between January 2008 and December 2019. Patient data was gathered from the hospital internal electronic database (AtlasMed), patient observation forms, surgery registry and the clinical imaging study database. We then divided the patients according to gender, age, and the histological type of lesion, based on the month of the year when their pathologies caused ICH. Inclusion criteria were the imaging and pathological confirmation of the vascular malformation, the presence of intracranial hemorrhage upon admission, as well as the surgical removal of the lesion. Patients with BVMs that did not present with rupture, were not confirmed as either AVM or CM by the pathologist, did not have a visible BVM on imaging study, or that did not benefit from surgery in order to have a final diagnosis were excluded from the study. Using Microsoft® Excel for Mac, we made the appropriate value distribution graphs. We performed Pearson’s chi square test to verify the relationship between season and rate of rupture of AVMs and CMs, individual month and rate 9 Does season influence the risk of haemorrhagic stroke from brain vascular malformations? of rupture, season and gender, and individual month and gender. RESULTS From the total of 202 BVMs operated in the specified interval by our senior neurosurgeon, 87 had ruptured prior to admission, out of which 71 (81.61%) were AVMs and 16 (18.31%) were CMs. There were 51 males (58.62%, with 40 AVMs and 11 CMs, representing 45.98% and 12.64% respectively) and 36 females (41,38%, with 31 AVMs and 5 CMs, accounting for 35.63% and 5.75% of cases respectively). The majority of ICH occurred during the months of July – 12 BVMs or 13.79% (10 AVVMs or 11.49%; 2 CMs or 2.3%); followed by December – 11 BVMs or 12.64% (10 AVMs or 11.49%; 1 CM or 1.15%). The months of March and November both accumulated 10 BVMs each (11.49%), the former having a total of 9 AVMs (10.34%) and 1 CM (1.15%), whereas the latter amassed 7 AVMs (8.05%) and 3 CMs (3.45%). Figure 1 reveals the total number of cases presenting with ruptured AVMs and CMs across each month. Using Pearson’s Chi squared test, we first compared AVMs to CMs in respects to the season they were most likely to rupture and obtained obtained a statistically significant correlation between the summer season and presentation with ruptured CM (chi2 value of 4.3291, p<.05). Moreover, when taking each month individually, we found a statistically significant correlation between August and ruptured CMs (chi2 value of 4.2902, p<.05) No other associations were statistically significant for this step. Tables 1 and 2 show the all of the results emerging from this group of investigations. Table 1. BVM Risk of Rupture According to Histological Type by Season and Period season/ period AVMs total/ month CMs total/ month chi2 value p value Interpre- tation Spring 19 1 1.905 8 1.9058 >.05 Non- Significant Summer 17 8 4.329 1 .03746 7 <.05 Significant Autumn 18 5 0.233 6 .62890 3 >.05 Non- Significant Winter 17 2 1.001 8 .31688 4 >.05 Non- Significant total 71 16 AVMs, arteriovenous malformations; CMs, cavernous malformations Table 2. BVM Risk of Rupture According to Histological Type by Month month AVMs total /month CMs total /month chi2 value P value Interpre tation January 5 1 0.0128 .910048 >.05 Non- Significant February 2 0 Could not be calculated March 9 1 0.53 .081428 >.05 Non- Significant April 6 0 Could not be calculated May 4 0 Could not be calculated June 4 3 3.0362 .096427 >.05 Non- significant July 10 2 0.0276 .868123 >.05 Non- significant August 3 3 4.2902 .038334 <.05 Significant Sept. 6 1 0.0855 .770012 >.05 Non- significant October 5 1 0.0128 .910048 >.05 Non- significant Nov. 7 3 0.2564 .612595 >.05 Non- significant Dec. 10 1 0.7256 .394304 >.05 Non- significant total 71 16 Figure 1. Diagram showing the cumulative incidence of ruptured brain vascular malformation based on type, according to calendar month. The ordinate shows the number of cases. AVM – arteriovenous malformation; CM – cavernous malformation. Next, we took into account only the patients with ruptured AVMs and compared the two genders in respects to rupture risks during all the same months combined. Figure 2 shows the distribution of ruptured BVMs for both sexes across each month. We found a statistically significant correlation between the male sex and presentation with a ruptured AVM in the month of December (chi2 value 10 Ioan Alexandru Florian, Teodora Larisa Timiș, Lehel Beni et al. of 5.3617, p<.05), whereas the female sex was significantly correlated with ruptured AVMs in the month of March (chi2 value of 4.8766, p<.05). Table 3 illustrates the complete results for this set of tests. Figure 2. Diagram showing the cumulative incidence of ruptured brain vascular malformation based on patient gender, according to calendar month. The ordinate shows the number of cases. VM – vascular malformation. Table 3. AVM Risk of Rupture According to Gender by Month month m. total /m. f. total /m. chi2 value P value Interpre- tation January 1 4 2.8874 .089277 >.05 Non- Significant Feb. 2 0 Could not be calculated March 2 7 4.8766 .027223 <.05 Significant April 4 2 0.2842 .593939 >.05 Non- significant May 3 1 0.6001 .438524 >.05 Non- significant June 1 3 1.6923 .193296 >.05 Non- significant July 5 5 0.1901 .662852 >.05 Non- significant August 1 2 0.6739 .411693 >.05 Non- Significant Sept. 4 2 0.2842 .593939 >.05 Non- significant Oct. 4 1 1.2243 .268522 >.05 Non- significant Novem ber 4 3 0.002 963931 >.05 Non- significant Dec. 9 1 5.3617 .020584 <.05 Significant total 40 31 AVMs, arteriovenous malformations; CMs, cavernous malformations Last, we considered all BVMs, regardless of type, and again compared the hemorrhage risk for the two sexes along all of the same months. We discovered a statistically significant correlation between the male sex and a hemorrhage from a vascular malformation of any kind in December (chi2 value of 5.412, p<.05), as well as the female gender and hemorrhagic stroke from any vascular malformation in the months January and August (chi2 value of 4.6763 equally for each month, p<.05). No other significant associations could be established. The complete results of this step can be viewed in Table 4. Table 4. Total BVM Risk of Rupture According to Gender by Month month m. total /m. f. total /m. chi2 value P value Interpretation Jan. 1 5 4.6763 .030582 <.05 Significant Feb. 2 0 Could not be calculated March 3 7 3.8155 .05078 >.05 Non-significant April 4 2 0.172 .678347 >.05 Non-significant May 3 1 0.4637 .495889 >.05 Non-significant June 4 3 0.0069 .934019 >.05 Non-significant July 7 5 0.0005 .982633 >.05 Non-significant Aug. 1 5 4.6763 .030582 <.05 Significant Sept. 5 2 0.5148 .473064 >.05 Non-significant Oct. 5 1 1.6225 .202741 >.05 Non-significant Nov. 6 4 0.0089 .925001 >.05 Non-significant Dec. 10 1 5.412 .019999 <.05 Significant total 51 36 DISCUSSIONS Within seasons themselves, the differences regarding the total number of ruptured AVMs was minor, whereas CMs tended to bleed more frequently during summer. The month of August showed a significantly higher relative frequency of ICH from CMs when compared to AVMs. Males showed a clearer tendency of BVM rupture in the month of December, whereas female patients had a higher propensity for ICH strictly from AVMs in March and from BVMs in general in January and August. Currently, there is a scarcity of studies dedicated to the effects of weather on the incidence of hemorrhagic stroke, especially those from preexisting vascular lesions such as aneurysms, AVMs or CMs. To the extent of our knowledge, a single other article addresses the influence of seasons on the risk of spontaneous hemorrhage from AVMs [6]. Hakan et al. discovered that the 11 Does season influence the risk of haemorrhagic stroke from brain vascular malformations? highest incidence for AVM rupture in Istanbul ensued during winter, while its lowest occurred in summer, yet there were no proven statistical correlations to substantiate their results. With 78 ruptured malformations spread across 20 years, the difference between hemorrhagic stroke from AVMs was also higher for men during autumn and winter, although these were not as considerable the ones in our series. The authors concluded that aside from seasonal weather variations and individual risk factors such as alcohol consumption, the relative increase of the population during winter may also play a role in the upsurge of hemorrhagic stoke during winter. Regarding cavernomas, our literature search did not yield any articles discussing the effect of meteorological conditions on their rupture patterns. Stroke mortality was proven to rise in holiday seasons, at least in certain regions such as in the Hiroshima Prefecture of Japan, or Taiwan [8, 14]. Yet the reason for this could not be accurately established due to a lack in more detailed clinical information like patient lifestyle, consumption of alcoholic beverages, smoking habits and so forth. In the Heilongjiang Province of China, the higher incidence of primary intracranial hemorrhage from hypertension within late spring and early autumn was attributed to the influence of daily mean ambient temperature, as well as its variation [22]. According to Zheng et al, unexpected changes in temperature, such as sudden drops during the hot weather or climbs during the cold, were met with an increase in the incidence of primary intracranial hemorrhage. Furthermore, in the same study, it was shown that the occurrence of SAH escalated throughout days with lower ambient temperature. On the other hand, in the Arab Peninsula, ischemic stroke was correlated with the higher amount of solar radiations specific for the summer season which could not be elucidated by physiological events indicative of either dehydration or hemoconcentration [19]. Therefore, the supposition that meteorological conditions influence the rates of stroke may not be far from true. The peaks occurring at the months of March, July and December might suggest a propensity for BVMs to rupture whenever extreme temperatures are reached, or when there are large thermic variations. The more frequent hemorrhages could also be attributed to the stress related to the estival season or the festive holidays and their respective activities. The significant difference between genders within early winter remains intriguing and enigmatic, yet it cannot be entirely attributed to weather effects alone. One should also take into account the behavioral and socioeconomic aspects of the patients in the studied population. As such, there are clear trends of increasing alcohol consumption in the form of spirits and wine during late autumn and early winter in our country, possibly due to the lower temperatures occurring in this period, but also as a consequence of increased stress correlated with a progressively dynamic socioeconomic environment, as well as a widening poverty gap [16, 17]. Although probable interference from other seasonal exposures such as dietary changes, hypovitaminosis or viral infections cannot be excluded. Studies have also tried to pinpoint specific individual meteorological factors which may influence the occurrence of spontaneous ICH. According to Neidert et al., relative humidity presented a strong fluctuation in the 2 days prior to ICH in patients with intracranial vascular lesions [15]. On the day of rupture relative humidity was significantly lower, suggesting that a low ambient humidity may promote hemodynamic and cardio- cerebrovascular conditions associated with ICH from these lesions. It has been shown that air humidity, along pollution and temperature, can influence the diameter of the brachial artery in patients with type 2 Diabetes mellitus, with higher temperature and humidity acting as vasodilators [21]. Conversely, it can be argued that a lower humidity may lead to vasoconstriction, although this might not suffice as a trigger for vascular malformation bleeds. Low humidity and air pressure cause an increase of blood viscosity via insensible water loss [7]. This in turn can also alter the autoregulation properties of arterial feeders in AVMs, as well as promote shear stress [13, 18]. Other reports revealed that low temperature is indeed a substantial risk factor for ICH, significantly correlated with the higher incidence of these occurrences during winter and early spring [11, 12]. Our findings are in line with these conjectures, showing that the increased incidence of ICH from vascular malformations during the colder months may not be coincidental. Nevertheless, it is as of yet unclear why so many our cases presented with rupture during summer and why the rate of rupture for CMs remains higher during the warmer months. 12 Ioan Alexandru Florian, Teodora Larisa Timiș, Lehel Beni et al. Nevertheless, these results should not be taken at face value. Statistical bias can occur from the relatively small number of cases from each individual month. We can, however, speculate that there is indeed a higher tendency for BVMs to rupture under certain environmental circumstances, such as high differences in temperature or atmospheric pressure. Yet determining these parameters retrospectively within a large timeframe can be difficult and more inaccurate than acquiring the same data upon hospital admission. Also, stress related to specific periods during the year may contribute to higher systemic blood pressure, ergo to an increase likelihood of these lesions to bleed. In order to resume and improve our research in the future, we intend to collect the meteorological data relevant to our patients on the same day in which they arrive. Another problem in our approach is that the total number of lesions is spread too thin across the twelve months, thus even a seemingly impressive frequency peak may actually denote a negligible difference. A larger case series, or a collaboration with other neurosurgical centers in this field, may solve this predicament in the future. Another limitation of this study is the lack of adequate meteorological data, particularly of the individual environmental factors such as temperature, relative humidity and air pressure in the days predating ICH. In a future continuation of this study, we propose a collaboration with the National Meteorological Association in order to obtain such relevant data. Despite this being a pilot study, we believe it could be further improved by enlarging the number of patients enrolled and by performing a multivariate analysis on the various individual meteorological, environmental, and geographical factors, as well as the individual characteristics of the patients and their BVMs. CONCLUSIONS According to our results, BVM rupture incidence is higher in the months of March, July and December. Moreover, CMs tend to rupture more frequently in the summer than AVMs. Males tend to present with rupture from AVMs during early winter, whereas females had ruptures in early spring. Despite the fact we achieved promising statistical results, due to the relatively low number of cases, our findings may not be applicable to a larger patient population. It may also be probable that our results are due to statistical bias, although we are optimistic regarding an actual relationship between meteorological conditions and an altered propensity of BVMs to rupture. This may also depend on geographical variations in risk factors. It seems likely that seasonal changes in the environment, especially extreme temperatures, might act as an additional risk factor for hemorrhagic stroke from vascular malformations, however these findings should be corroborated with supplementary case series from other centers, or a large prospective trial. CONFLICTS OF INTEREST The authors have no conflicts of interest to report. REFERENCES 1. Bertalanffy H, Burkhardt JK, Kockro RA, Bozinov O, Sarnthein J. Resection of cavernous malformations of the brainstem. Cavernous Malformations Nerv Syst. 2011:143-160. doi:10.1017/CBO9781139003636.016. 2. Choudhri O, Chen RP, Bulsara K. 21 - Cavernous Malformations of the Brain and Spinal Cord. Fourth Edition. Elsevier Inc. doi:10.1016/B978-0-323-43140- 8.00021-4. 3. Çevik Y, Doğan NÖ, Daş M, Ahmedali A, Kul S, Bayram H. The association between weather conditions and stroke admissions in Turkey. Int J Biometeorol. 2015 Jul;59(7):899-905. doi: 10.1007/s00484-014-0890-9. Epub 2014 Aug 22. PMID: 25145443. 4. Florian IS, Barițchii A, Trifoi SV. [Arterio-venous Mallformations]. In: Popescu I, Ciuce C, Editors. Tratat de Chirurgie Vol VI. 2nd edition. Editura Academiei Române. 2014; 5: 402-410 [Text in Romanian]. 5. Florian IS, Perju-Dumbravă L. [Therapeutic Options in Hemorrhagic Strokes]. Editura Medicală Universitară „Iuliu Hațieganu” Cluj-Napoca. 2007; 2.1: 331-346 [Text in Romanian]. 6. Hakan T, Kizilkilic O, Adaletli I, Karabagli H, Kocer N, Islak C. Is there any seasonal influence in spontaneous bleeding of intracranial aneurysm and and/or AVM in Istanbul?. Swiss Med Wkly. 2003;133(17-18):267-272. 7. Hashiguchi N, Takeda A, Yasuyama Y, Chishaki A, Tochihara Y. Effects of 6-h exposure to low relative humidity and low air pressure on body fluid loss and blood viscosity. Indoor Air. 2013 Oct;23(5):430-6. doi: 10.1111/ina.12039. Epub 2013 Mar 28. PMID: 23464811. 8. Huang HK, Chang WC, Hsu JY, et al. Holiday Season and Weekend Effects on Stroke Mortality: A Nationwide Cohort Study Controlling for Stroke Severity. J Am Heart Assoc. 2019;8(8):e011888. 9. Iliescu BF, Poeata I. [Cerebral Cavernomas]. In: Popescu I, Ciuce C, Editors. Tratat de Chirurgie Vol VI. 2nd edition. 13 Does season influence the risk of haemorrhagic stroke from brain vascular malformations? Editura Academiei Române. 2014; 5: 402-410 [Text in Romanian]. 10. Kalb S, Gross BA, Nakaji P. Vascular Malformations (Arteriovenous Malformations and Dural Arteriovenous Fistulas). In: Ellenbogen RG, Sekhar LN, Kitchen ND, Editors. Principles of Neurological Surgery. 4th edition. Elsevier. 2018; 20: 313-324. 11. Kim HJ, Kim JH, Kim DR, Kang HI, Moon BG, Kim JS. Age and meteorological factors in the occurrence of spontaneous intracerebral hemorrhage in a metropolitan city. J Cerebrovasc Endovasc Neurosurg. 2014 Sep;16(3):209-15. doi: 10.7461/jcen.2014.16.3.209. Epub 2014 Sep 30. PMID: 25340022; PMCID: PMC4205246. 12. Magalhães R, Silva M, C, Correia M, Bailey T. Are Stroke Occurrence and Outcome Related to Weather Parameters? Results from a Population-Based Study in Northern Portugal. Cerebrovasc Dis 2011;32:542-551. doi: 10.1159/000331473. 13. Morgan M, Winder M. Haemodynamics of arteriovenous malformations of the brain and consequences of resection: a review. J Clin Neurosci. 2001 May;8(3):216-24. doi: 10.1054/jocn.2000.0795. PMID: 11386794. 14. Nakaguchi H, Teraoka A. Relationship between the occurrence of spontaneous intracerebral hemorrhage and holidays and traditionally unlucky days in Fukuyama City, Hiroshima Prefecture, Japan. J Stroke Cerebrovasc Dis. 2007 Sep-Oct;16(5):194-8. doi: 10.1016/j.jstrokece rebrovasdis.2007.04.002. PMID: 17845915. 15. Neidert MC, Sprenger M, Mader M, Esposito G, Hosp JA, Bozinov O, Regli L, Burkhardt JK. A High-Resolution Analysis on the Meteorological Influences on Spontaneous Intracerebral Hemorrhage Incidence. World Neurosurg. 2017 Feb;98:695-703.e19. doi: 10.1016/j.wneu.2016.12.006. Epub 2016 Dec 10. PMID: 27965078. 16. Petrovici DA, Ritson C. Food consumption patterns in Romania. British Food Journal. 2000; 102(4):290 – 308. 17. Popescu A. Changes and Trends in Wine Production and Consumption in the World and Romania During the Period 2007-2018. Scientific Papers Series Management, Economic Engineering in Agriculture and Rural Development. 2019; 19(2): 345-362. 18. Rossitti, S., Svendsen, P. Shear stress in cerebral arteries supplying arteriovenous malformations. Acta neurochir. 1995; 137, 138–145. https://doi.org/10.1007/BF0218 7185. 19. Salam A, Kamran S, Bibi R, et al. Meteorological Factors and Seasonal Stroke Rates: A Four-year Comprehensive Study. J Stroke Cerebrovasc Dis. 2019;28(8):2324-2331. 20. Wang K, Li H, Liu W, You C. Seasonal variation in spontaneous intracerebral hemorrhage frequency in Chengdu, China, is independent of conventional risk factors. J Clin Neurosci. 2013;20(4):565-569. doi:10.1016/j.jocn.2012.02.052. 21. Zanobetti A, Luttmann-Gibson H, Horton ES, Cohen A, Coull BA, Hoffmann B, Schwartz JD, Mittleman MA, Li Y, Stone PH, de Souza C, Lamparello B, Koutrakis P, Gold DR. Brachial artery responses to ambient pollution, temperature, and humidity in people with type 2 diabetes: a repeated-measures study. Environ Health Perspect. 2014 Mar;122(3):242-8. doi: 10.1289/ehp.1 206136. Epub 2014 Jan 7. PMID: 24398072; PMCID: PMC3948021. 22. [22] Zheng Y, Wang X, Liu J, Zhao F, Zhang J, Feng H. A Community-Based Study of the Correlation of Hemorrhagic Stroke Occurrence with Meteorologic Factors. J Stroke Cerebrovasc Dis. 2016;25(10):2323-2330.