Organ involvement occurs in all forms of hereditary haemolytic anaemia

The life expectancy of patients with hereditary haemolytic anaemia (HHA) has shown remarkable improvement. The availability of transfusions, chelation therapy and other supportive treatment options has improved life expectancy well into adulthood (Elmariah et al, 2014; Vitrano et al, 2017). HHA comprises a heterogeneous group of genetic diseases characterized by premature destruction of red cells. It is the number one cause of anaemia burden in the high income Western World and, due to immigration and better survival of patients, it is a growing health problem (Kassebaum et al, 2014). For example: sickle cell disease (SCD) accounts for 300·000 newly diagnosed infants each year globally, and this number is expected to increase to 400·000 by 2050 (Piel et al, 2013). Many patients are confronted with chronic organ involvement as they get older. Disease-related organ involvement is an important determinant of morbidity and prognosis in SCD and β-thalassaemia. (Powars et al, 2005; Fitzhugh et al, 2015; Vitrano et al, 2017). However, for other forms of HHA, little is known about disease-related organ involvement, especially for non-transfusion dependent patients. This study aimed to make an inventory of the occurrence of organ involvement in HHA, encompassing (early markers of future) organ damage, markers of altered organ function (e.g. endocrine changes) and other non-haematological symptoms known to influence morbidity and prognosis in chronically ill patients (e.g. inflammation). This could be a first step towards determining whether organ involvement screening protocols would be necessary for all patients with HHA. This is a cross sectional, observational study. Patients were participants of the ZEbRA-study (Netherlands Trial Register [NTR] identifier, NTR5337), in the University Medical Center Utrecht, the Netherlands. Adult patients with HHA, with and without transfusion history, were eligible to enrol. Data was collected through chart review. For details see Table SI. Only patients with >10 items available were included in the analysis. Statistical analysis was performed using non-parametric tests because of the abnormal distribution of some of the data. One hundred patients were enrolled between 2016 and 2017. Ninety had information on organ involvement available and were included in the analysis. Patients had a median age of 40 years (range 18–84) and a haemoglobin concentration (Hb) of 111 g/l (range 64–164) and 46 patients (51%) were female. Patients were divided into four disease categories: SCD (15 patients), other haemoglobin disorders (12 β-thalassaemia, 4 unstable haemoglobins, 1 HbH-disease variant), enzyme disorders (23 pyruvate kinase deficiency, 4 glucose-6-phosphate-dehydrogenase deficiency, 2 hexokinase deficiency, 1 glutamate cysteine ligase deficiency), and membrane disorder (21 hereditary spherocytosis, 5 hereditary xerocytosis, 2 hereditary elliptocytosis, Table 1). Seventy-eight patients (87%) showed at least some form of organ involvement. Patients who did not show organ involvement had less clinical test results available than patients who did (median 17 vs. 24, P < 0·001). Most reported forms of organ involvement were iron overload (41/53, 77%) vitamin D deficiency (32/54, 59%) and cholecystectomy (39/90, 43%). Patients who did have organ involvement did not show differences in Hb (110 vs. 126 g/l, P = 0·220) or reticulocytes (236 vs. 202 × 109/l, P = 0·879) compared to patients who did not. Of the 31 patients that had never received transfusions (SCD excluded), 24 (77%) showed at least some form of organ involvement. Most reported forms of organ involvement were iron overload (65%), vitamin D deficiency (58%), and cholecystectomy (39%). (Table 2) There was no difference in iron loading between male and female patients. Never-transfused patients with organ involvement did not show differences in Hb (132 vs. 135 g/l, P = 0·755) or reticulocytes (256 vs. 173 × 109/l, P = 0·600) compared to never-transfused patients without organ involvement. Organ involvement occurred in all disease categories, but there was a difference in number of positive items (as percentage of the total amount of scored items, P = 0·031). Patients with SCD scored a median of 4 positive items out of 24 scored items. For patients with other haemoglobin disorders this was 3/26, for enzyme disorders 3/23 and for membrane disorders 1/19 (Table 2). Our study shows that organ involvement occurs in all forms of HHA included in this study and is not exclusive to patients with SCD or β-thalassaemia. Notably, many forms of organ involvement were not related to clinical symptoms, this is in line with previous observations in SCD. (van Beers et al, 2008; van Tuijn et al, 2017). These results emphasise that organ involvement is not limited to frequently transfused patients. Using a stringent definition of “never-transfused patients”, 77% showed organ involvement. This is relevant, given that these patients are often considered to be mildly affected and might not regularly attend the clinic. The observational set-up of the study makes it impossible to draw conclusions on causality. As an example, the high percentage of vitamin D deficiency in this cohort is similar to that found in the healthy Dutch population (Keyzer et al, 2014). However, as our study also identifies osteoporosis in 26% of patients, evaluating vitamin D levels is very relevant in our patients. Further, our data shows that patients with organ involvement have more clinical test results, such as magnetic resonance imaging, available. This represents information bias, but also indicates that screening for organ involvement does not structurally occur in patients with HHA. In SCD, although genetic and haematological modifiers of disease severity are well described at the population level, no specific characteristics to identify a high risk profile for organ involvement has been identified at the patient level (Piel et al, 2017; van Tuijn et al, 2017). We, too, were unable to identify a clinical or haematological profile that discriminates patients with organ involvement from those without. Therefore, it is important to implement systematic screening guidelines for organ involvement in patients with HHA. In summary, our study shows that organ involvement is not limited to patients with SCD or β-thalassaemia, but also occurs in other forms of HHA. Importantly, organ involvement also occurs in never-transfused patients. Following the improved survival in HHA, the unfavourable effect of organ involvement on morbidity and mortality is an important new concern. Screening for organ involvement, which has a high prevalence and treatment options available, seems prudent. We suggest this could include screening for iron overload, osteoporosis, vitamin D and sex hormone deficiency and microalbuminuria. S.S. performed data collection, analysed results, created the figures. S.S., R.W. and E.J.B. prepared the manuscript, J.V. and S.H. contributed to data collection, S.S., R.W., E.J.B., J.V., S.H., R.S. and W.S contributed to the revision of the manuscript. The authors declare no competing financial interests. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.