Post-COVID syndrome and type 2 diabetes mellitus in Kazakhstan: clinical manifestations and vaccine efficacy
Article information
Abstract
Purpose
In this article, we report the results of a survey investigating post–coronavirus disease 2019 (COVID-19) syndrome in patients with type 2 diabetes mellitus and the impacts of vaccination on long-term manifestations. From February 2022 to April 2023, a survey of patients with type 2 diabetes and people without diabetes who were treated for a coronavirus infection was conducted in Kazakhstan.
Methods
Participants were invited via social media to voluntarily participate in this study. A total of 417 surveys were included in this study, comprising 212 patients with type 2 diabetes and 205 without diabetes. We compared persistent complaints after recovery in patients with and without diabetes mellitus (DM), as well as vaccination status.
Results
The results of this study on self-reported symptoms of prolonged COVID show that more than half of those infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in Kazakhstan have at least one persistent symptom after recovery. Based on the list of prolonged COVID-19 symptoms reported by patients with type 2 DM (T2DM), exertional dyspnoea, fatigue, respiratory discomfort, headaches, and sleep disturbances are among the most common ongoing conditions, and in many cases last more than 3 months after COVID-19.
Conclusions
Patients with T2DM experience more severe and prolonged symptoms than those without diabetes. Additionally, vaccination lowers COVID-19 hospitalization risk and decreases the need for adjusting hypoglycemic therapy, such as insulin treatment, after recovering from SARS-CoV-2 infection.
Highlights
· This study investigates post–coronavirus disease 2019 (COVID-19) syndrome in patients with type 2 diabetes and the impact of vaccination on long-term symptoms.
· Vaccination significantly reduced hospitalization rates and the need for hypoglycemic therapy corrections in type 2 diabetes mellitus patients.
· The study highlights the prevalence and severity of prolonged COVID-19 symptoms in Kazakhstan.
Introduction
According to the World Health Organization, a total of 763 million cases of coronavirus disease 2019 (COVID-19) were registered from the beginning of the COVID-19 pandemic until April 19, 2023, of which 6.9 million or about 1% of the cases were fatal [1]. As of April 2023, 1,498,668 confirmed cases of COVID-19 were registered in Kazakhstan, of which 13,843 were fatal, which also accounts for about 1% of cases [2]. At the same time, as of January 24, 2023, the number of people vaccinated with component 1 is 10,858,101, and 10,629,063 people have been fully vaccinated. Despite this, data on the course of COVID-19 obtained during the pandemic indicate the emergence of a new pathological condition defined as "post-COVID-19 syndrome" or "prolonged COVID" (post-COVID syndrome [PCS]) [3].
British scientists were the first to describe PCS. Since then, systematic reviews, meta-analyses, and cohort studies have expanded as the number of patients with persistent coronavirus infection increased. These studies define PCS as comprising ongoing physical, medical, and cognitive effects after contracting COVID-19, including lasting immunosuppression and lung, heart, and blood vessel fibrosis [4]. A distinctive feature of PCS is the emergence of new symptoms that can change over time. A number of mechanisms have been proposed to explain this [5,6].: (1) deficient immune response, which allows the virus to keep replicating longer; (2) systemic damage from excessive inflammatory reaction or immune system changes (cytokine storm syndrome); (3) physical or mental/psychosocial consequences (anxiety, depression, post-traumatic stress disorder, effects of lockdown or social isolation); (4) reinfection with the same or a different variant of SARS-CoV-2.
A meta-analysis of 21 studies revealed that 80% of patients with SARS-CoV-2 who have one and/or more long-term symptoms after recovery may suffer from a variety of as many as 55 long-term effects [7]. The following 5 were the most common: fatigue (58%), headache (44%), attention disorder (27%), hair loss (25%), and dyspnoea (24%) [8]. According to some researchers, female sex, high blood pressure, overweight (body mass index above 30 kg/m2), and diabetes are among the main risk factors provoking a long course of coronavirus infection [9-12]. The combination of coronavirus infection and type 2 diabetes mellitus (T2DM) directly affects morbidity, mortality, and the occurrence of prolonged and relapsing courses of both coronavirus infection and associated PCS [13,14].
According to official data (National Register of Diabetes Mellitus of the Republic of Kazakhstan) [15], the number of patients with T2DM increased from 208,000 to 380,000 from 2019 to 2022 in Kazakhstan, and continues to grow. Incomplete data for 2023 indicates that the number is greater than 400,000. The number of young patients with T2DM, including women, has increased, which increases the rehabilitation period of the employable part of patients in the reproductive period. Considering these facts, health care programs for T2DM are the most expensive, and according to our data, the consequences of prolonged post-COVID conditions are directly related to the increase in the number of patients with T2DM. At the time of study, the available data did not allow us to reliably determine how effective vaccines were in relation to the consequences of coronavirus infection in patients with T2DM [15-17]. However, it was found that timely and correctly vaccinated patients with T2DM get sick less often and tolerate the disease more easily, and although complications occurred in some cases, mortality rates were still higher in patients who did not receive vaccines. It has also been shown that the effectiveness of vaccines decreases over time.
Based on previous findings, the purpose of the current research was to study the main symptoms of PCS in patients with T2DM, as well as to determine the effects of the presence, absence, or partial vaccination on the occurrence and course of PCS in patients with T2DM, differing by sex, age and concomitant diseases.
Materials and methods
We conducted an online questionnaire among patients with confirmed coronavirus infection complicated by PCS using an online survey platform that was distributed through social networks. The information sheet had the following form and equivalent information in all 3 language versions: "You are invited to take part in this study because: you had COVID-19 (or have not yet cleared up, or there are residual symptoms in the form of the following symptoms: respiratory discomfort, exertional dyspnoea, dyspnoea at rest…); you are 18 or older."
Among respondents who reported having symptoms of prolonged COVID-19, 62.6% (n=261) were women, 44.4% (n=185) were aged 18–44, and 55.6% (n=232) were from 45 to 70 years of age, of which only 17.5% (n=73) of patients were vaccinated. In addition, 50.8% (n=212) were patients with T2DM; of these, 23.6% (n=50) had diabetes of duration less than 5 years, 29.2% (n=62) from 6 to 10 years, 31.1% (n=66) from 11 to 19 years, and 16% (n=34) for more than 20 years. Most respondents (63.3%, n=264) received outpatient treatment, and 36.7% (n=153) received inpatient treatment (Table 1).
All respondents were informed about the anonymity of the questionnaire before participating in the study. The survey responses contained no personal information, and the email addresses collected to distribute the survey were encrypted as anonymous participant identifiers. The data were collected from February 2022 to April 2023. The questionnaires were presented in Kazakh, Russian, and English, included 47 questions, and took 30–40 minutes to complete. The study was approved by the Research Ethics Committee at the Asfendiyarov Kazakh National Medical University. (IRB No. E-15236) A total of 435 questionnaires were collected. After a preliminary check, 18 questionnaires were excluded due to incomplete data and a lack of information indicating the presence of certain symptoms. In the final report, we analyzed the responses of 417 participants who underwent coronavirus infection. Of these, 212 patients had T2DM, while 205 were not suffering from diabetes.
The study variables included: age; sex; concomitant chronic diseases, including T2DM; vaccination status and dose number, as well as the interval between vaccinations; concomitant infections; whether the patient was hospitalized or treated on an outpatient basis; correction of hypoglycemic therapy after infection; prevalence of symptoms in 10 organ systems; and presence, duration, and frequency of these symptoms.
Statistical analyses were performed using IBM SPSS Statistics ver. 19.0 (IBM Co., Armonk, NY, USA). Pearson chi-square test was applied to test the relationships between nominal qualitative variables, while the relationships between variables such as sex, age, the presence of T2DM, vaccination status, the number of vaccines administered, and the presence of certain complaints were analyzed by chi-square test for independence of variables. Statistical significance was accepted at P<0.05. Confidence intervals were also calculated at the level of 95% of the means and proportions.
Results
1. Frequency of symptoms of prolonged COVID-19
In this study we assessed the incidence of symptoms of prolonged COVID-19 and the duration and frequency of these symptoms. In the total sample (n=417), we identified 4,144 events grouped by 41 symptoms characterizing 10 human organ systems. The most common self-reported symptoms were fatigue or decreased exercise tolerance in 53.5% (95% confidence interval [CI], 48.4%–58%); exertional dyspnoea in 49.4% (46.5%–56.1%); and respiratory discomfort (inspiratory restriction, feeling of heaviness in the chest) in 49.4% (44.6%–53.7%). The least common included ear pain in 3.8% (2.2%–5.8%) and oliguria in 4.8% (2.6%–7.0%) (Fig. 1).
Most of the common symptoms were similar in men and women. The exception is that the following prevailed in women: pelvic control disorder (7.3% vs. 1.9%, P=0.018; odds ratio [OR], 4.004; 95% CI, 1.100–13.700), headache (47.5% vs 35.9%, P=0.021; OR, 1.616; 95% CI, 1.070–2.400), joint pain (37.9% vs 28.2%, P=0.043; OR, 1.556; 95% CI, 1.010–2.300) and symptoms of trouble or anxiety (28.4% vs. 19.2%, P=0.038; OR, 1.662; 95% CI, 1.020–2.600) (Table 2).
Symptom prevalence varied significantly with age. In individuals aged 45–70, symptoms linked to respiratory, cardiovascular, and nervous systems were twice as frequent as in younger people. These include respiratory discomfort, exertional dyspnoea, dyspnoea at rest, cough, chest pain, heart pain, headaches, sleep disturbance, numbness, paresthesia, and limb weakness. Age also significantly affects diarrhea, weight loss, polyuria, joint pain, muscle pain, tinnitus, smell disorder, and fever, all of which were more common in older individuals [18]. These findings emphasize age-related differences in symptom manifestation among COVID-19 patients. On the contrary, skin rash was more common in the young group, amounting to 11.4% vs. 6.9%, P=0.112, however, the significance was not reliable due to the small number of observations that noted this symptom. Fatigue/impaired exercise tolerance, which is a common symptom in the general population, also did not vary significantly by age (56.9% vs. 49.2%, P=0.117) (Table 3, Fig. 2).
Table 3 presents a comprehensive overview of symptom prevalence across different age groups. Notable findings include the higher prevalence of respiratory discomfort, shortness of breath on exertion, cough, and chest pain among the 45–70 age group compared to the 18–44 age group. Additionally, cardiovascular symptoms such as chest tightness and heart pain, as well as several neurological symptoms, including numbness, paresthesia, and weakness in the limbs, were more common in the older age group. Conversely, younger individuals reported a higher frequency of skin manifestations, specifically noise in ears and smell disorder, as well as a greater likelihood of experiencing temperature increases. These findings highlight age-related differences in symptom presentation among the surveyed individuals, providing valuable insights into the varied experiences of different age cohorts in relation to their health and symptoms.
According to the duration of the manifestation of PCS, we found that in 31% of patients, complaints persisted from 3 to 6 months, in 32% of patients from 2 to 3 months, in 26% of patients from 1 to 2 months, in 8% of patients to 4 weeks, and the smallest number (3%) of patients noted the persistence of complaints from 6 months to 1 year. In addition, there is a significant relationship between the duration of these complaints with the presence of DM in patients (P=0.02) and with the presence of vaccination (P<0.001), but the relationship was not determined by sex (P=0.1) or age of respondents (P=0.2).
We found that 58.5% of patients had complaints on a daily basis, and 41.5% noted that these symptoms bothered them periodically. Reliable differences in the frequencies of these complaints were found according to the presence of T2DM (63.7% vs. 53.2%, P=0.03) and vaccination (60.3% vs. 41.1%, P<0.001). On the contrary, there were no significant differences by age (60.8% vs. 55.7%, P=0.294) or by sex (57.5% vs. 60.3%, P=0.5).
2. Features of prolonged COVID-19 in patients with T2DM
Among all respondents with T2DM (n=212), the most frequent symptom was exertional dyspnoea in 61.8% (range, 55.2%–68.9%), and among patients without diabetes (n=206), the most frequent symptom was fatigue/decreased exercise tolerance in 46.3% (range, 39%–53.5%). Furthermore, symptoms of prolonged COVID-19 were more common in patients with T2DM than those without. For instance, in the diabetes group, exertional dyspnoea and reduced mental capacity were significantly more frequent, with chest tightness and various other symptoms also showing notable differences. Conversely, some symptoms, like sore throat, loss of taste, and skin rash, had similar prevalence in both groups. Notably, polyuria and nocturia were markedly more frequent in the diabetes group, while oliguria was less common. These findings highlight the distinct impact of diabetes on COVID-19 symptoms (Table 4, Fig. 3).
Table 4 illustrates the disparities in symptom prevalence between patients with T2DM and those without DM. T2DM patients are more likely to experience symptoms across multiple systems. Notably, they are more prone to respiratory discomfort, shortness of breath on exertion, cough, chest pain, and several cardiovascular symptoms, including chest tightness and palpitations. Neurological symptoms such as headaches, sleep disturbances, numbness in the lower limbs, paresthesia, and weakness in the limbs are also more prevalent in T2DM patients. Additionally, T2DM patients report higher frequencies of gastrointestinal symptoms, urinary issues, musculoskeletal discomfort, cognitive and emotional problems, and ENT symptoms. Overall, T2DM patients tend to experience a broader range of symptoms compared to those without DM.
Vaccinated patients comprised only 17.5% of the total population, of which 70% were vaccinated with vector, and 30% with inactivated vaccines. At the same time, patients who received only 1 dose accounted for only 9.6% (n=7) and all 7 of those patients (100%) fell ill with coronavirus infection within 4 weeks. The majority (43.9%) of patients who received full vaccination became infected 6 months after the start of vaccination, 27.3% fell ill after 3 months, 16.7% fell ill up to 4 weeks, and the remaining 7.6% and 4.5% became infected after 1 month and after 2 months, respectively. In general, fatigue/decreased exercise tolerance was the most common symptom in this group, which made up 28.8%, when among unvaccinated patients, this figure was 58.7% (P<0.001; OR, 3.522; 95% CI, 2.032–6.107). At the same time, the prevalence of symptoms of PCS was twice as high among respondents who were infected before receiving the vaccine as among those vaccinated (Table 5, Fig. 4).
Table 5 displays symptom prevalence in vaccinated and unvaccinated patients. Notable differences exist between the 2 groups across various systems. For instance, symptoms related to respiratory discomfort, exertional dyspnoea, and cough were significantly more common in unvaccinated patients. Similar disparities were observed in cardiovascular and nervous system symptoms. Gastrointestinal symptoms, including abdominal pain, nausea, diarrhea, decreased appetite, and aversion to food, were more prevalent in unvaccinated individuals. Additionally, symptoms of joint pain, muscle pain, cognitive issues, depression, and anxiety were more widespread among unvaccinated patients. Hair loss and fatigue were also more common in unvaccinated individuals. However, temperature increase was less frequent in vaccinated patients.
In vaccinated patients with T2DM, the prevalence and duration of all symptoms were also significantly lower than in unvaccinated patients with DM. The most common symptoms among vaccinated patients with diabetes were fatigue (35.3%), headaches (32.4%) and lack of taste (32.4%), and among unvaccinated patients with diabetes, the most common symptoms were exertional dyspnoea (69.1%), fatigue (65.2%), and respiratory discomfort (60.1%). In addition, the effect of vaccination on the frequency of hospitalizations for coronavirus infection and the correction of hypoglycemic therapy were determined. The majority (82.2%, P<0.001; OR, 3.100; 95% CI, 1.600–5.900) of vaccinated patients did not seek inpatient care, and received outpatient treatment. The proportion of unvaccinated patients who required correction of hypoglycemic therapy (including the appointment of insulin therapy) was 2 times higher (67.6% vs. 32.4%, P=0.012; OR, 2.800; 95% CI, 1.200–6.100) than among vaccinated patients with T2DM.
The survey findings of 417 coronavirus patients showed that PCS has various consequences depending on the initial state of the immune system, as well as concomitant diseases that often affect several organ systems. This kind of study is the first conducted in the territory of Kazakhstan. We were able to identify 41 symptoms from the data (4,144 invariants) characterizing changes in several organs and organ systems that cause health problems not only in physical, but also in cognitive and emotional aspects. This analysis showed that female, middle-aged, or elderly patients with complicated T2DM, as well as patients who were infected prior to vaccination, had more symptoms associated with PCS.
Discussion
According to the frequency of occurrence, the symptoms of PCS were distributed as follows: fatigue or decreased exercise tolerance (53.5%); exertional dyspnoea (51.3%); respiratory discomfort (49.4%); and hair loss (44.1%). These findings are similar to those of large population-based studies that rank fatigue, breathing problems, and hair loss among the most common post-COVID-19 symptoms [19,20]. The data confirm previous conclusions that patients have persistent neuropsychiatric symptoms, however, in these studies, attention and concentration difficulties were the most common symptoms, whereas in our survey, headaches (43.2%) and sleep disorders (41.7%) prevailed [21]. According to sex, the symptoms were distributed as follows: among women, the most common were pelvic control disorder (P=0.018, OR=4.004, CI 1.100– 13.700), headache (P=0.021; OR, 1.616; 95% CI, 1.070–2.400), joint pain (P=0.043; OR, 1.556; 95% CI, 1.010–2.300), and symptoms of trouble/anxiety (P=0.038; OR, 1.662; 95% CI, 1.020–2.600), which correspond to the results obtained by other researchers [22-25]. These differences may be due to the fact that women are more involved in such studies and are more aware of their health than men.
We found that middle-aged and elderly patients in Kazakhstan reported more symptoms of PCS than young people, especially a large number of complaints were observed from the respiratory, cardiovascular and nervous systems. The same results were obtained by Italian studies via comparison of the prevalence and duration of symptoms in adults and children living in the same household, which found that the spectrum of symptoms of PCS in adults is wider, the duration is longer, the course is more severe, and that many adults had signs of multisystem damage [26]. This may be due to the fact that middle-aged and elderly patients have more concomitant diseases than children. There are studies that found the opposite, however; studies in Ecuador and the UK have shown that the risk of developing PCS increases with decreasing age [27].
An important conclusion of the present study is that we found correlations between the prevalence and duration of symptoms in patients with T2DM. Similar results were obtained in previous studies [28]. Thus, the presence of concomitant diseases is a risk factor for the development of PCS. According to our data, the most common symptom among patients with T2DM is exertional dyspnoea (61.8% vs. 40.5%, P<0.001), whereas in the control group it was fatigue/decreased exercise tolerance (46.3%). Moreover, the proportion of patients with T2DM reporting symptoms that make it possible to classify COVID-19 as persistent (in each case) is 2–4 times higher in almost all systems than in patients without DM. These symptoms include, in particular, decreased mental capacity (P<0.001; OR, 2.400; 95% CI, 1.600–3.760), chest tightness (P=0.005; OR, 2.100; 95% CI, 1.200–3.400), polyuria (P<0.001; OR, 5.100; 95% CI, 2.700– 9.500), manifestations of nocturia (P<0.001; OR, 3.700; 95% CI, 2.100-6.400), and symptoms of trouble/anxiety (P<0.001; OR, 2.400; 95% CI, 1.300–3.500). Perhaps these differences are due to the fact that most patients with DM have concomitant diseases of the cardiovascular system and complications of diabetes itself from the nervous system, and the manifestations of polyuria and nocturia are a consequence of glucose toxicity.
Despite this, a number of symptoms did not significantly differ in prevalence between patients with and without DM, in particular, sore throat (10.4% vs. 10.2%, P=0.9), lack of taste (42.9% vs. 40.5%, P=0.68), and skin rash (9.4% vs. 8.3%, P=0.8). Spanish researchers reported that there was no significant relationship between diabetes and the severity of PCS among the patients they studied [29]. With a degree of certainty, symptoms could be classified as markers characteristic only of patients with T2DM, but the prevalence and duration of individual symptoms were significantly higher among patients with T2DM in the present study. However, an Indian study found that patients with T2DM who had coronavirus infections complicated by PCS exhibited greater fatigue than those who did not have COVID-19 [30]. Currently, there are data on the duration and frequency of PCS only in the general population, but without distinguishing patients with T2DM [27]. Therefore, in the present study we tried to determine the relationships between the presence of T2DM in patients who underwent coronavirus infection complicated by PCS and the duration and frequency of symptoms. We found that 63.7% of patients with T2DM complained of daily manifestations of symptoms, 71.2% of patients have symptoms lasting up to 3 months, and 28.8% have symptoms lasting from 3 months to a year. Moreover, the duration (P=0.02) and frequency (P=0.03) of these complaints were significantly associated with the presence of T2DM in patients who experienced coronavirus infection complicated by PCS.
Currently, studies conducted around the world have supported the effectiveness of vaccination against coronavirus infection, expressed in decreases in the incidence rate, frequency of hospitalization, and mortality from COVID-19, but there are no data on the effects of vaccination on the prevalence and duration of PCS in patients with diabetes in general and with T2DM in particular [31-33]. We found that respondents who received the 2 recommended vaccinations had lower prevalence of all symptoms than respondents who were not vaccinated. Among all vaccinated patients, the predominant symptoms were fatigue (28.8%), exertional dyspnoea (27.4%), and respiratory discomfort (26%). The majority (82.2%, P<0.001) of vaccinated patients were treated on an outpatient basis, and it can be assumed that the course of infection was mild or asymptomatic. These findings coincide with those of a number of previous studies [34].
As for the effectiveness of vaccination in patients with T2DM, many studies have concluded that vaccination against COVID-19 should be a priority in this group of patients [35-37]. Nevertheless, at the beginning of the pandemic, some patients with diabetes were observed to mount inadequate humoral responses to SARS-CoV-2. These patients may have received vaccinations while already infected with coronavirus. In vaccinated patients with T2DM, the prevalence and duration of all symptoms were also significantly lower than in unvaccinated patients with DM. Among the most frequent symptoms in vaccinated patients with T2DM were fatigue (35.3%), headaches (32.4%), and lack of taste (32.4%); among the unvaccinated the most frequent symptoms were exertional dyspnoea (69.1%), fatigue (65.2%), and respiratory discomfort (60.1%). Moreover, the proportion of unvaccinated patients who required correction of hypoglycemic therapy (including the appointment of insulin therapy) was 2 times higher (P=0.012; OR, 2.800; 95% CI, 1.200-6.100) than among vaccinated patients with type 2 DM. This trend may be due to the fact that the course of infection was mild or asymptomatic in most vaccinated patients with T2DM, so glucocorticosteroids were not required in these patients.
This is the first study of the long-term effects of COVID-19 in patients with T2DM in Kazakhstan. The results of this study of self-reported symptoms of prolonged COVID show that more than half of those infected with SARS-CoV-2 in Kazakhstan had at least one persistent symptom after recovery. Among the list of prolonged COVID-19 symptoms reported by patients with T2DM, exertional dyspnoea, fatigue, respiratory discomfort, headaches, sleep disturbances were among the most common ongoing conditions, and in many cases, they lasted more than 3 months after COVID-19 infection.
This research has some limitations. We relied on patients reporting their own symptoms instead of using medical tests to confirm findings objectively. Furthermore, our sample size was relatively small, which could impact the results and whether they may be generalized to other contexts. More studies are needed to understand the clinical and immunological characteristics of post-COVID syndrome in people with type 2 diabetes who were infected by COVID-19. It also seems promising to investigate how diabetes medications affect post-COVID syndrome development, since there are many options but no guidance regarding which drugs are best for patients recovering from infection. Overall, additional research including more patients and objective clinical data would strengthen our understanding of post-COVID syndrome in the type 2 diabetes population. Examining how glycemic control medications could influence post-COVID outcomes would also be useful, given the range of drug options without evidence about priorities for post-infection care. In addition, the social status of patients deserves separate additional study for all groups of patient. Moreover, more research is needed on preventive measures for diabetic patients, in particular, the priority of vaccine selection for these groups, as well as the frequency of vaccinations. Since the protection provided by a particular type of vaccine has limited duration, its effectiveness is correlated with time, from 6 months to 1 year.
In conclusion, we found that middle-aged and elderly patients, women, patients with T2DM, and unvaccinated patients are more susceptible to the occurrence of PCS with a larger number of complications and a more severe, prolonged course. Vaccination significantly reduces the risk of disease, while if disease occurs, vaccination reduces the number of hospitalizations for COVID-19. Vaccination also reduces the need for correction of hypoglycemic therapy, including insulin therapy after recovery from SARS-CoV-2, which reduces the overall cost of managing patients and improves quality of life.
Notes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Funding
This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Data availability
The data that support the findings of this study can be provided by the corresponding author upon reasonable request.
Author contribution
Conceptualization: SA, ZA; Data curation: SA, SB; Formal analysis: ZA, SB, AA; Methodology: SB, TD, AA; Project administration: SA, ZA; Visualization: SB, TD, AA; Writing - original draft: SA, ZA, SB, TD, AA; Writing - review & editing: SA, ZA, SB, TD, AA