This retrospective cohort study was conducted in the Namazi hospital, a major referral center for stroke in Shiraz with a large catchment area in the Fars province (southern Iran). The study periods are defined as a) pre-COVID period [21th March 2018 to 20th March 2019] and b) COVID-19 period [20th March 2020 to 20th March 2021]. Iran reported its first confirmed case of COVID-19 on 19th February 2020 [17]. The National COVID-19 vaccination program was started after the recruitment time. [17] Accordingly, no recruited patient was vaccinated. As all official medical and demographic reports are presented in the Iranian calendar, which starts on 21st March, the above-mentioned dates were selected for the study periods [18].
According to the Statistical Center of Iran, the estimated > 20-year-old population of Fars province in the first and second period was 3,487,000 and 3,547,000 people, respectively, based on the National Population and Housing Census results in the year 2016 [19]. Namazi hospital covers all CVST patients from the metropolitan area of Shiraz and severe CVST patients from other parts of Fars province. The referral system from the catchment area had no change between the pre-COVID period and the COVID period. Urban and rural areas are defined according to the Statistical Center of Iran [19].
Definitions
All patients with age>20year-old with the definite final diagnosis of CVST were included in our study. The diagnosis criteria of CVST was based on the presence of relevant clinical symptoms (headache, focal neurological syndromes, mental status disturbance, etc.) and radiological assessments of the brain (Computed Tomography -CT-, CT venography, Magnetic resonance imaging -MRI- or MR venography) [20]. Patients with incomplete medical records as well as patients with indefinite diagnosis were excluded.
COVID-19 attributed investigations were done for patients if any clinical relevance was found. COVID-19 is considered in patients with positive reverse transcriptase-polymerase chain reaction (RT-PCR) from the oro- or naso-pharynx [21]. Patients who developed COVID-19 infection as a nosocomial infection was excluded.
Disability or death at the time of discharge and three-month follow-up was assessed according to the modified Rankin Scale (mRS) score. Poor outcome is defined as mRS ≥ 3 at discharge and three-month after the index event.
The precipitating risk factors of CVST are grouped in the following categories: Sex-specific risk factors -pregnancy, puerperium and oral contraceptive pills-, malignancy, infection, thrombophilia, hematologic, rheumatologic, mechanical, dehydration, miscellaneous and unrecognized.
The superficial sinus venous system was defined as superior and inferior sagittal, transverse, sigmoid, occipital, and cavernous sinuses and all attributed superficial veins (Trolard, Labbe, superficial middle cerebral vein, etc.). The deep system was defined as straight sinus, the vein of Galen, the basal vein of Rosenthal, internal cerebral veins, and all attributed veins (septal, thalamostriate, etc.).
Inclusion and exclusion criteria, variables, and outcomes
We entered all data from patients diagnosed with CVST from the time of admission into a newly organized registry system designed by the engineering team of the Neurology Research Center of Shiraz University of Medical Sciences (Iran ministry of health registry code: 9001013381). To ensure full entry of patients in the registry system, using the international classification of diseases 10th version (ICD-10), we searched all medical records with ICD-10 diagnostic codes of G08 for “Intracranial and intraspinal phlebitis and thrombophlebitis”, O87.3 for “Cerebral venous thrombosis in the puerperium”, I63.6 for “Cerebral infarction due to cerebral venous thrombosis, nonpyogenic”, I67.6 for “Non-pyogenic thrombosis of the intracranial venous system" and I61.9 for "Non-traumatic intracerebral hemorrhage”. All patients aged more than 20 with the final diagnosis of CVST were included in the registry system after reconfirmation of the diagnosis by a qualified vascular neurologist. Patients with indefinite diagnosis were excluded.
The following information was collected: demographic data including age, sex, and risk factors; mode of the onset of symptoms (classified as acute: <48 hours, sub-acute: 48 hours to 30 days and chronic: >30 days from symptom onset to hospital entry); any neurologic symptoms and signs, Imaging of brain including computed tomography (CT), magnetic resonance imaging (MRI), MR cerebral venography (MRV) and CT cerebral venography (CTV), mRS and Glasgow Coma Scale (GCS) scores, laboratory data including vasculitis panel and hypercoagulability tests and treatments including anticoagulants.
All patients were investigated by routine laboratory tests and some specific laboratory tests, including vasculitis panel and hypercoagulability tests, to find the underlying causes of CVST listed in the supplement file. In the COVID-19 period, the SARS-CoV-2 RT-PCR test is performed in clinically suspected patients.
All patients with CVST received therapeutic doses of intravenous heparin or subcutaneous enoxaparin. Based on the neurologist’s decision, anticoagulation was continued with warfarin or direct oral anticoagulants. All data, including mortality data, were updated at the time of booked follow-up outpatient neurology clinic visits.
Statistical analysis
Analysis was performed on the data using IBM SPSS Statistics for Windows, Version 16.0. Armonk, NY: IBM Corp. The P-value<0.05 is considered significant in all analyses.
The crude hospitalization rate per 1,000,000 population in the two periods was the dependent variable. The quantitative variables are shown using mean ± standard deviation (SD) or median with interquartile range (IQR) according to the distribution pattern. The qualitative variables are reported using numbers with percentages (%). A Chi-Square test was used to compare the hospitalization rate change between pre-covid-19 and COVID-19 periods. Besides, independent variables (e.g., predisposing risk factors and main demographic variables) were compared using Chi-Square tests. Due to the non-normally distributed pattern of our data, we applied the Mann-Whitney test to compare the age of participants between pre-COVID19 and COVID-19 periods. In addition, using Cox Proportional Hazards regression analysis, we compared the adjusted Hazard ratio (aHR) of mortality between mentioned periods. To evaluate the trend of crude hospitalization rate of CVST, aggregated monthly hospitalization of CVST was used.
Bayesian interrupted time series was used to examine the change of hospitalization rate across the study period. The BSTS package was used to perform time series analysis in the R 4.1 environment. The final time series model was fitted with adjusting the effects of age and male to female ratio variables. The 95% Credible Intervals (CrI) were used to report the Bayesian time series model results.
Study Protocol approval
The institutional review board and the Ethics Committee of Shiraz University of Medical Sciences (SUMS) approved the study protocol(IR.SUMS.REC.1399.098). This study follows the ethical standards of the institutional and national research committee and with the Helsinki Declaration or comparable ethical standards [22]. Data can be shared with other centers upon the approval of the Ethics Committee.