Clinical feature
Among the 41 patients, 10 were males and 31 were females, aged (54.4 ± 15.9) years old(range from 15 to 81)years. The male age was (61.6 ± 17.2) years and the female age was (52.0 ± 15.1) years. The most common clinical manifestations were anemia (97.6%),dizziness or fatigue(92.7%), jaundice(85.4%), hepatosplenomegalgia (53.7%), soy colored urine (26.8%), melena (19.5%), fever (12.2%), abdominal pain and vomiting (4.9%), as shown in Fig. 1.
Laboratory examination
The median HB of 41 patients at admission was 59.0 (48.0, 70.5) g/L, with 14 cases(35.0%) of moderate anemia, 24 cases (60.0%),of severe anemia.and 2 cases (5.0%) of extremely severe anemia. In terms of cell morphology, there were 25 cases (62.5%) of positive cell anemia, 13 cases (32.5%) of large cell anemia and 2 cases (5.0%) of small cell anemia. Five patients (12.2%) were combined with thrombocytopenia, the lowest being 23*109/L. Leukopenia was observed in 9 patients (22.0%). Leukocyte and platelet decreased simultaneously in only 1 case. Bone marrow aspiration was performed in 36 patients, and bone marrow reports indicated hyperplastic anemia in 27 cases (75.0%), inverted grain/red ratio in 8 cases (22.2%), and hypoplasia in 1 case (2.8%). Ret count increased by 0.244 (0.195, 0.299) in 40 patients (97.6%). Only 1 patient had a decreased Ret count. There were 39 patients (95.1%) with varying degrees of LDH increase of 411.0 (289.0, 607.5) U/L, 35 patients (85.4%) with TBIL increase of 53.4 (36.6, 84.8) umol/L, mainly indirect bilirubin increase. Serum Fer increased 407.0 (220.9, 519.5) ug/L in 38 patients (92.7%).
Diagnostic methods
In the diagnosis of AIHA, it is observed that there are symptoms and signs of anemia, hemolysis, and the presence of autoantibodies. Peripheral blood and bone marrow aspiration examinations that indicate retiocytosis and cytoplasmic hyperplasia are also indicative, and specific diagnostic criteria depend on positive direct or indirect anti-human globulin test (Coomb,s)[9]. During this study, 41 patients were diagnosed with AIHA for the first time, and 33 (80.5%) were Coomb,s positive. Three patients (7.3%) tested negative for Coomb,s after the initial screening, yet their clinical symptoms strongly indicated the presence of AIHA. Coomb,s were positive results upon re-examination at the blood station located downtown.Five cases (12.2%) were diagnosed as Coomb,s negative after repeated re-examination. However, the clinical symptoms remained consistent, and other forms of hemolytic anemia were ruled out. The patients responded well to glucocorticoid therapy, confirming the diagnosis of Coomb,s negative AIHA.
A total of 19 patients (46.3%) were diagnosed with primary autoimmune hemolytic anemia (AIHA), while 22 patients (53.7%) were diagnosed with secondary AIHA. Among the secondary factors, 3 patients (13.6%) were diagnosed with Hashimoto's thyroiditis, 5 patients (22.7%) were diagnosed with systemic lupus erythematosus, rheumatoid arthritis was diagnosed in 2 patients (9.1%), aplastic anemia was diagnosed in 1 patient (4.5%), tumor was diagnosed in 1 patient (4.5%), and secondary infection was diagnosed in 10 patients (45.5%). Specifically, 3 patients of urinary tract infection,3 patients of pulmonary infection, 2 patients of hepatitis virus infection, 1 patient of cytomegalovirus infection, and 1 patient of intestinal infection were documented (refer to Table 1).
Prognosis
In this research, glucocorticoid alone (specifically methylprednisolone) was administered to 34 patients, accounting for 82.9% of the total. A combination of glucocorticoid and gamma globulin shock therapy was administered to 12.2% of the patients. Additionally, 2 patients (4.9%) received a combination of glucocorticoids, gamma globulin as well as plasmapheresis. Furthermore, 56.1% of the patients received transfused rinsed red blood cells.Based on the therapeutic outcomes observed in the patients, a total of 24 individuals (accounting for 58.5%) experienced either complete or partial alleviation of clinical symptoms were classified into the group with good prognosis. Conversely, the group with poor prognosis consisted of 17 patients (representing 36.6%) who did not exhibit any improvement following treatment and subsequently passed away during their hospital stay(with 2 patients died).
Analysis of prognostic factors There were no statistically significant differences observed WBC, PLT, LDH, ALT, TBIL, IBIL, Bun, Cr and Fer levels between the two groups (P > 0.05). However, the HB level 62.0 g/L in the good prognosis group (range from 49.3 to 79.3 g/L) was significantly higher than that in the poor prognosis group 47.0 g/L(range from 30.2 to 58.5 g/L) (P < 0.001). Additionally, the CRP level 3.9 mg/L in the good prognosis group ( range from 1.7 to 13.2 mg/L) was significantly higher than that in the poor prognosis group 32.8 mg/L ( range from 26.8 to 49.3 mg/L) (P = 0.001).
The mean AST level in the good prognosis group was 17.4 ± 6.7 U/L, whereas the mean AST level in the poor prognosis group was 25.5 ± 12.1 U/L. This difference in mean values was found to be statistically significant (P = 0.019). Furthermore, the good prognosis group had 7 patients (29.2%) with secondary AIHA, whereas the poor prognosis group had 15 patients (88.2%) with secondary AIHA. This difference in proportions was also found to be statistically significant (P < 0.001). Please refer to Table 2. HB, CRP, AST, and secondary etiology were incorporated as covariates in a binary logistic regression analysis to examine their association with the prognosis of AIHA. The results revealed that a decrease in HB levels and the presence of secondary etiology were identified as independent risk factors for a poor prognosis of AIHA. Please refer to Table 3 .
ROC curve and area under the curve (AUC) were described and compared.
Meaningful indicators in multivariate Logistic regression analysis were calculated: the area under the ROC curve of HB and secondary etiology were 0.762 (95% confidence interval: 0.616–0.909) and 0.795 (95% confidence interval: 0.653–0.938). The results presented in Table 4 and Fig. 2 indicate that HB exhibited the highest level of specificity and sensitivity towards secondary etiology.
Misdiagnosis situation
In the present investigation, a total of 9 individuals were found to have been misdiagnosed during their initial diagnosis, resulting in a misdiagnosis rate of 22.0%.Among the patients included in the study, a total of 8 individuals (19.5%) were initially diagnosed incorrectly with upper gastrointestinal hemorrhage, while one patient (2.4%) was misdiagnosed as having acute cholecystitis.In this particular cohort, a total of 8 patients presented with "melena" as their primary complaint, accompanied by positive fecal occult blood and a decline in hemoglobin levels. Given the suspicion of upper digestive tract bleeding, appropriate measures were taken in accordance with the emergency diagnosis and treatment protocol for acute upper digestive tract bleeding[10–11] after admission. During the blood preparation procedure, the blood transfusion department conducted cross-matching tests on two patients, which revealed the presence of autoantibodies.In light of the potential occurrence of hemolysis, a comprehensive battery of hemolysis examinations was conducted, revealing a positive result in the anti-human globulin test (Coomb's test), thereby confirming the presence of AIHA.Six patients were administered a standard red blood cell suspension upon admission. Among these patients, 4 exhibited elevated levels of TBIL, IBIL, and LDH compared to their pre-transfusion levels. Additionally, 2 patients experienced fever during the transfusion process.
After complete hemolysis examination, AIHA was confirmed. A patient presented with the chief complaint of experiencing abdominal pain accompanied by vomiting. An abdominal ultrasound examination revealed the presence of gallstones along with inflammation of the gallbladder, known as cholecystitis. Consequently, the patient was admitted to the hospital ward for further management of gallstones with cholecystitis.Upon admission, the blood test revealed the presence of macrocytic anemia. Further investigation into the underlying causes of this anemia led to the indication of AIHA. Among the 9 patients who were misdiagnosed, 1 patient was male and 8 patients were female. Four patients had a poor prognosis while 5 patients had a good prognosis. Seven patients were diagnosed with secondary AIHA and 2 patients were diagnosed with primary AIHA. The mean age of the patients was 54.7 ± 14.7 years. The average HB level was 59.0 (40.0, 64.0) g/L. The average time of misdiagnosis was 2.8 days (range from 1 to 7 days). In terms of laboratory examination, the average TBILlevel was 49.1 (31.6, 67.3) umol/L, LDH level was 464.0 (334.5, 638.5) U/L, CRP level was 5.9 (4.3, 99.5) mg/L, Fer level was 415.9 (232.7, 667.7) ug/L, and Ret level was 0.236 (0.209, 0.255). The mean levels of HB, LDH, TBIL, CRP, Ret, and Fer in the 9 patients were not statistically significant when compared to the entire group. Please refer to Table 5.