Since the beginning of the COVID-19 pandemic, concerns have been expressed worldwide for patients with hemoglobinopathies and their vulnerability to SARS-CoV-2 infection. Data from Lebanon confirmed a role of underlying comorbidities on COVID-19 severity, but no deaths among a cohort of thalassemia patients.¹ Patients with sickle cell disease (SCD) displayed a broad range of severity after SARS-CoV-2 infection, spanning from a favorable outcome unless pre-existing comorbidities (UK cohort)² to high case mortality in US.³ History of pain, heart, lung, and renal comorbidities was identified as risk factors of worse COVID-19 outcomes by the US SECURE-SCD Registry.⁴ While Italy experienced a death rate in the general population among the highest in the world, preliminary data from the first wave of the pandemic showed a lower than expected number of infected thalassemia patients (updated up to April 10, 2020), likely due to earlier and more vigilant self-isolation compared to the general population.⁵

To explore the vulnerability to SARS-CoV-2 infection, the Italian Society for Thalassemia and Hemoglobinopathies (SITE) designed a study to compare the prevalence and mortality of COVID-19 in individuals with hemoglobinopathies and the general Italian population (EMO AER COVID-19 study). The study was approved by Institutional Review Board authorities, registered on clinicaltrials.gov (NCT04746066), and was performed in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. Designed to gather data from multiple healthcare providers in Italy, it allowed for collecting relevant demographics and clinical data on a dedicated electronic Case Report Form (eCRF) (available at https://covid19.site-italia.org) by each participating center.

We enrolled patients with transfusion-dependent thalassemia (TDT), nontransfusion-dependent thalassemia (NTDT), and sickle cell disease (SCD) referred to participating centers and diagnosed with SARS-CoV-2 infection in the study period March 6, 2020 to April 7, 2021. SARS-CoV-2 infection was confirmed by either a positive swab of the upper or lower respiratory tract or serology. Patients with less than 15 days of follow-up from either the onset of symptoms or a SARS-CoV-2 positive test were excluded. Twenty-nine centers from 13 Italian Regions participated in the EMO AER COVID-19 study. These centers regularly provide care for approximately 6200 patients with hemoglobinopathies (3400 TDT, 1500 NTDT, 1300 SCD), representing 65% of the Italian population affected by these pathologies. Therefore, this sample is highly representative of Italian patients with hemoglobinopathies followed by an organized and widespread national network, providing both high coverage and high definition of data.

During the 398-day study period, a total of 345 SARS-CoV-2 infections were recorded (overall, prevalence 5.5%): 230 cases among TDT (prevalence 6.8%), 50 among NTDT (prevalence 3.3%), and 65 among SCD patients (prevalence 5.0%). In the SCD group, 49% of patients were β-Thal/HbS. Diagnosis of COVID-19 was confirmed by a positive swab in 91% of the cases and by the presence of serum IgG in 9% of the cases. Among reported cases, 52% were female. The median age at the infection was 41 years (IQR: 29–48, range: 0.75–85), with 10% of patients being pediatric (median age: 8 years, IQR: 4–11). Seventy-four percent of patients had at least one comorbidity at the time of infection. The most common were: splenectomy or functional asplenia (50%), iron overload (23%), liver disease (19%), heart disease (16%), and diabetes (8%). ABO blood groups were distributed as follows: 50% were O, 33% were A, 15% were B, and 2% of patients were AB. We observed a broad spectrum of COVID-19 severity, ranging from no symptoms (83/345, 24%) to severe manifestations (66/345, 19%) and death (7/345, 2%). The most common symptoms were fever (157/345, 46%), cough (145/345, 42%), fatigue and diffuse pain (119/345, 34%), and anosmia and ageusia (104/345, 30%). Severe symptoms, such as difficulty breathing or thoracic pain, affected 62/345 (18%) of patients; 55/345 (16%) had pneumonia; and 1 patient experienced pulmonary thromboembolism. Overall, 68 (20%) patients required hospitalization, 15 (8 TDT, 2 NTDT, 5 SCD) in high-intensity care units (ICU). Nine out of 68, all with pneumonia (1 TDT, 1 NTDT, and 7 SCD), required additional or ad hoc blood transfusions due to acute hemoglobin drop. The median hospitalization time was 11 days (IQR: 5–21, range: 1–102 days, information available for 46 patients).

Seven patients experienced fatal COVID-19 during the period of observation: 4 TDT (46/M, 48/M, 49/M, 56/F), 1 NTDT (45/M), 2 SCD (57/M, 57/F), both with the diagnosis of β-Thal/HbS. One TDT patient (57/F) and two patients (1 SCD, 52/M; 1 TDT, 57/F) died after the conclusion of the analysis and were not included in this survey. The overall lethality rate was 2.0%. The age-standardized lethality ratio (SLR) was then calculated as the ratio between the observed and the expected number of deaths, based on the age-specific rates in the Italian-COVID population. The resulting SLR was 4.8 (±3.5, 95% CI). All the fatal episodes were observed starting from November 2020. For hospital admission, age was a risk factor in TDT (OR = 1.03; CI: 1–1.1; p = .04) and NTDT (OR = 1.05; CI: 1–1.1; p = .04), but not in SCD. In TDT only, the presence of underlying lung or heart disease increased the risk to be admitted to the hospital (OR = 4.5, CI = 1.1–19.3, p = .04; OR = 2.9, CI = 1.0–8.0, p = .04). For the SCD group, chronic liver disease was associated with a higher risk of hospital admission (OR = 7.5, CI: 1.1–53.5, p = .04). For ICU admission and mortality, the presence of previous pulmonary disease was a risk factor only for TDT (OR = 5.6, CI: 1.2–25.1, p = .03; OR = 26.6; CI: 2.3–311.4; p = .01, respectively).

According to our results, the prevalence of COVID-19 in hemoglobinopathies in Italy was similar to the general population (5.5% vs. 6.2%) in the first 13 months of the pandemic.

Considering the known underestimation of SARS-CoV-2 prevalence in the Italian population and the greater reliability of the same estimation in our strictly monitored patients, we speculate that the risk of infection in hemoglobinopathies was actually reduced. This hypothetical difference should be explained by the effectiveness of early recommendations from dedicated healthcare providers and the prudent attitude of the chronic patients in front of risk, as already reported from expert centers in other countries.⁶

The estimation of lethality is complex: 95.6% of the confirmed COVID-19 deaths in the Italian population have occurred in subjects in ages 60 or greater and 86.2% of the deaths in ages 70 or greater. Lethality rates for COVID-19 infected patients were 26.7% for ages 90 years or greater, 19.8% for ages 80–89 years, 9.4% for ages 70–79 years, and 2.7% for ages 60–69 years.

Our study population is significantly younger in age overall, with only 1.4% subjects infected above 70 years of age, reflecting the age-distribution of the hemoglobinopathies in Italy. The proper comparator for our population is the segment of the Italian population younger than 60 years of age, which experienced 5% of the total COVID-19 deaths, with lethality rates varying from <0.1% (age 20–29 years) to 0.6% (age 50–59 years). While no significant differences were observed in patients aged 0–30s, significantly higher lethality was observed in subjects aged 40–49 and 50–59 years, where all fatal cases were registered. Assuming for hemoglobinopathies the same lethality rates of Italians with comparable age, the number of observed deaths in hemoglobinopathies is approximately 5-fold the expected one (Figure 1). All deaths occurred in patients in the fourth to fifth decades of life, mostly obese, splenectomized, and with numerous comorbidities. Surprisingly, none of them (except one for which no recent clinical data are available) had a significant iron overload. Both deaths in the SCD group occurred in patients with β0-Thal/HbS, while there were no fatal events among patients with homozygous HbS, in agreement with the local genotype distribution that is characterized by a high prevalence of older Caucasian β-Thal/HbS patients (homozygous HbS are more frequent among younger patients).

Age was a risk factor for hospital admission due to SARS-CoV-2 infection in both TDT and NTDT, but not in SCD. Other risk factors were the presence of underlying comorbidities at the time of infection, particularly chronic lung, heart, or liver disease. In addition, chronic lung disease was a significant risk factor for ICU admission or mortality (in TDT only).

The main limitations of this work are represented by the evaluation of indirect outcomes of COVID-19 severity; in addition, not all the Italian centers taking care of these patients were involved in the study. However, the data presented here include the large majority of known patients affected by hemoglobinopathies in Italy. Another limitation of our study is the inability to consider the effects of early vaccination in this at risk cohort compared with the general population.

Our data clearly indicate that patients affected by hemoglobinopathies have up to a five times higher likelihood of suffering lethal SARS-CoV-2. Thus, these patients should be referred to specific and expert healthcare providers. Future studies should monitor the long-term effect of COVID-19 in patients with hemoglobinopathies. More relevantly, the effectiveness of vaccines should be evaluated in these patients to address the presence of any possible difference with the general population.

自 COVID-19 大流行开始以来，全世界都对血红蛋白病患者及其对 SARS-CoV-2 感染的脆弱性表示担忧。来自黎巴嫩的数据证实了潜在合并症对 COVID-19 严重程度的影响，但在一组地中海贫血患者中没有死亡。¹镰状细胞病 (SCD) 患者在感染 SARS-CoV-2 后表现出广泛的严重程度，从良好的结果（除非预先存在的合并症）（英国队列）²到美国的高病例死亡率。³疼痛史、心脏、肺和肾脏合并症被美国 SECURE-SCD 登记处确定为 COVID-19 结局较差的危险因素。⁴虽然意大利的总人口死亡率位居世界前列，但第一波大流行的初步数据显示，受感染的地中海贫血患者人数低于预期（更新至 2020 年 4 月 10 日），这可能是由于早先与一般人群相比，更加警惕的自我隔离。⁵

为了探索 SARS-CoV-2 感染的脆弱性，意大利地中海贫血和血红蛋白病协会 (SITE) 设计了一项研究，以比较 COVID-19 在血红蛋白病患者和意大利普通人群中的患病率和死亡率 (EMO AER COVID-19学习）。该研究得到了机构审查委员会当局的批准，在临床试验.gov (NCT04746066) 上注册，并按照良好临床实践指南和赫尔辛基宣言进行。它旨在从意大利的多个医疗保健提供者那里收集数据，它允许每个参与中心在专用的电子病例报告表 (eCRF)（可在 https://covid19.site-italia.org 上获得）上收集相关的人口统计数据和临床数据。

我们招募了在 2020 年 3 月 6 日至 4 月的研究期间转诊至参与中心并被诊断为 SARS-CoV-2 感染的输血依赖性地中海贫血 (TDT)、非输血依赖性地中海贫血 (NTDT) 和镰状细胞病 (SCD) 患者2021 年 7 月 7 日。SARS-CoV-2 感染通过上呼吸道或下呼吸道的阳性拭子或血清学证实。从出现症状或 SARS-CoV-2 检测呈阳性后随访不到 15 天的患者被排除在外。来自意大利 13 个地区的 29 个中心参与了 EMO AER COVID-19 研究。这些中心定期为大约 6200 名血红蛋白病患者（3400 TDT、1500 NTDT、1300 SCD）提供护理，占受这些疾病影响的意大利人口的 65%。所以，

在为期 398 天的研究期间，共记录了 345 例 SARS-CoV-2 感染（总体而言，患病率 5.5%）：TDT 中 230 例（患病率 6.8%），NTDT 中 50 例（患病率 3.3%）和 65 例SCD 患者（患病率 5.0%）。在 SCD 组中，49% 的患者为 β-Thal/HbS。COVID-19 的诊断通过 91% 的病例拭子阳性和 9% 的病例存在血清 IgG 得到证实。在报告的病例中，52% 为女性。感染时的中位年龄为 41 岁（IQR：29-48，范围：0.75-85），10% 的患者是儿童（中位年龄：8 岁，IQR：4-11）。74% 的患者在感染时至少有一种合并症。最常见的是：脾切除术或功能性无脾（50%）、铁过载（23%）、肝病（19%）、心脏病（16%）和糖尿病（8%）。ABO血型分布如下：50%为O型，33%为A型，15%为B型，2%为AB型。我们观察到 COVID-19 的严重程度范围很广，从无症状 (83/345, 24%) 到严重表现 (66/345, 19%) 和死亡 (7/345, 2%)。最常见的症状是发烧（157/345, 46%）、咳嗽（145/345, 42%）、疲劳和弥漫性疼痛（119/345, 34%）以及嗅觉丧失和听力丧失（104/345, 30%） . 62/345 (18%) 的患者出现呼吸困难或胸痛等严重症状；55/345 (16%) 患有肺炎；1例患者出现肺血栓栓塞。总体而言，68 名（20%）患者需要住院治疗，15 名（8 名 TDT，2 名 NTDT，5 名 SCD）在高强度护理病房（ICU）。68 人中有 9 人患有肺炎（1 例 TDT、1 例 NTDT 和 7 例 SCD），由于急性血红蛋白下降，需要额外或临时输血。

7 名患者在观察期间出现了致命的 COVID-19：4 名 TDT（46/M、48/M、49/M、56/F）、1 名 NTDT（45/M）、2 名 SCD（57/M、57/ F)，两者都诊断为 β-Thal/HbS。一名 TDT 患者（57 岁/F）和两名患者（1 名 SCD，52/M；1 名 TDT，57/F）在分析结束后死亡，未纳入本次调查。总体致死率为2.0%。然后，根据意大利 COVID 人群的年龄特异性死亡率，将年龄标准化致死率 (SLR) 计算为观察到的死亡人数与预期死亡人数之间的比率。得到的 SLR 为 4.8 (±3.5, 95% CI)。从 2020 年 11 月开始观察到所有致命事件。对于住院，年龄是 TDT（OR = 1.03；CI：1-1.1；p  = .04）和 NTDT（OR = 1.05；CI：1-1.1）的危险因素; p = .04)，但不在 SCD 中。仅在 TDT 中，潜在的肺或心脏病的存在增加了入院的风险（OR = 4.5，CI = 1.1-19.3，p  = .04；OR = 2.9，CI = 1.0-8.0，p  = . 04)。对于 SCD 组，慢性肝病与更高的入院风险相关（OR = 7.5，CI：1.1-53.5，p  = .04）。对于 ICU 入院和死亡率，既往肺部疾病的存在仅是 TDT 的危险因素（OR = 5.6，CI：1.2-25.1，p  = .03；OR = 26.6；CI：2.3-311.4；p  = .01，分别）。

根据我们的结果，在大流行的前 13 个月，意大利血红蛋白病中 COVID-19 的患病率与普通人群相似（5.5% 对 6.2%）。

考虑到意大利人群中 SARS-CoV-2 流行率的已知低估以及我们严格监测的患者中相同估计的更高可靠性，我们推测血红蛋白病感染的风险实际上降低了。正如其他国家的专家中心已经报告的那样，这种假设的差异应该由专门的医疗保健提供者的早期建议的有效性和慢性病患者在风险面前的谨慎态度来解释。⁶

致死率的估计很复杂：意大利人口中 95.6% 的确诊 COVID-19 死亡发生在 60 岁或以上的受试者中，86.2% 的死亡发生在 70 岁或以上的受试者中。90 岁或以上年龄的 COVID-19 感染患者的致死率为 26.7%，80-89 岁为 19.8%，70-79 岁为 9.4%，60-69 岁为 2.7%。

我们的研究人群总体年龄显着年轻，只有 1.4% 的受试者在 70 岁以上感染，这反映了意大利血红蛋白病的年龄分布。适合我们人口的比较对象是 60 岁以下的意大利人口，他们经历了 COVID-19 死亡总数的 5%，致死率从 <0.1%（20-29 岁）到 0.6% 不等（年龄 50-59 岁）。虽然在 0-30 岁的患者中没有观察到显着差异，但在 40-49 岁和 50-59 岁的受试者中观察到显着更高的致死率，所有死亡病例均已登记。假设血红蛋白病的致死率与年龄相近的意大利人相同，观察到的血红蛋白病死亡人数约为预期死亡人数的 5 倍（图 1）。所有死亡均发生在 40 岁至 50 岁的患者中，大多为肥胖、脾脏切除和多种合并症。令人惊讶的是，它们中没有一个（除了没有最近的临床数据可用的）有明显的铁过载。SCD 组的两例死亡均发生在 β0-Thal/HbS 患者中，而在纯合 HbS 患者中没有致命事件，这与以老年高加索人 β-Thal/HbS 患病率高为特征的局部基因型分布一致患者（纯合 HbS 在年轻患者中更常见）。

在 TDT 和 NTDT 中，年龄是因 SARS-CoV-2 感染而入院的危险因素，但在 SCD 中则不然。其他风险因素是感染时存在潜在的合并症，特别是慢性肺、心脏或肝脏疾病。此外，慢性肺病是入住 ICU 或死亡的重要危险因素（仅在 TDT 中）。

这项工作的主要局限性体现在评估 COVID-19 严重程度的间接结果；此外，并非所有照顾这些患者的意大利中心都参与了这项研究。然而，这里提供的数据包括意大利绝大多数已知的受血红蛋白病影响的患者。我们研究的另一个限制是，与普通人群相比，无法考虑早期疫苗接种对这个高危人群的影响。

我们的数据清楚地表明，受血红蛋白病影响的患者患致命性 SARS-CoV-2 的可能性高达五倍。因此，应将这些患者转诊给特定和专业的医疗保健提供者。未来的研究应监测 COVID-19 对血红蛋白病患者的长期影响。更相关的是，应在这些患者中评估疫苗的有效性，以解决与一般人群可能存在的差异。