Objective:To investigate the efficacy and safety of combining venetoclax (VEN) with hypomethylated drugs (HMA) in the treatment of higher-risk (IPSS-R score >3.5) myelodysplastic syndromes (MDS) .Methods:From March 2021 to December 2022, forty-five MDS patients with intermediate and high risk were treated with VEN in combination with HMAs. Clinical data were collected and analyzed retrospectively, including gender, age, MDS subtype, IPSS-R score, treatment regimen, and efficacy, etc. Kaplan-Meier method and Cox regression model were used to analyze univariate and multivariate of survival prognosis.Results:①Forty-five patients with MDS, including ninety-one percent were classified as high or very high risk. According to the 2023 consensus proposal for revised International Working Group response criteria for higher-risk MDS, the overall response rate (ORR) was 62.2% (28/45), with the complete response rate (CR) was 33.3% (15/45). For twenty-five na?ve MDS, the ORR was 68% (17/25) and the CR rate was 32% (8/25). In nonfirst-line patients, the ORR and CR were 55% (11/20) and 35% (7/20) respectively. The median cycle to best response was 1 (1-4). ②With a median followup of 189 days, the median overall survival (OS) time was 499 (95% confidence interval, 287-711) days, and most patients died from disease progression. Responders had a significantly better median OS time than nonresponders (499 days vs 228 days, P<0.001). Multifactor analysis revealed that IPSS-R score and response to treatment were independent prognostic factors for OS; the presence of SETBP1 gene mutations was associated with a longer hospital stay (51.5 days vs 27 days, P=0.017) . Conclusions:There is clinical benefit of venetoclax in combination with hypomethylated agents in patients with higher-risk MDS, but adverse events such as severe hypocytopenia during treatment should be avoided.

OBJECTIVE: To explore effectiveness of positive support reduction and internal fixation in the treatment of femoral neck fractures. METHODS: A clinical data of 74 patients with femoral neck fractures treated with hollow screw internal fixation between September 2017 and September 2021 was retrospectively analyzed. Based on the quality of fracture reduction, they were divided into positive support reduction group (group A, n=25), negative support reduction group (group B, n=21), and anatomical reduction group (group C, n=28). There was no significant difference in baseline data such as gender, age, cause of injury, disease duration, fracture side, Garden classification, and fracture line position classification between groups (P>0.05). The occurrence of complications such as early fixation failure, femoral neck shortening, non-union of fractures, and femoral head necrosis in three groups, as well as the Harris score of the hip joint were recorded and compared. RESULTS: All patients had primary healing of incisions after operation and were followed up more than 12 months. The follow-up time for groups A, B, and C was (21.1±5.7), (22.6±4.3), and (21.9±4.1) months, respectively; there was no significant difference between groups (P>0.05). There was no significant difference in the incidences of non-union of fractures, early internal fixation failure, and the femoral head necrosis between groups (P>0.05). The incidence and length of femoral neck shortening, and the hip Harris score at last follow-up in groups A and C were all superior to those in the group B, with significant difference (P<0.05). There was no significant difference in the above indicators between groups A and C (P>0.05). CONCLUSION Positive support reduction can provide a good biomechanical environment for the healing of femoral neck fractures, thereby achieving a higher fracture healing rate, reducing the occurrence of femoral neck shortening, minimizing the function of hip joint, and achieving effectiveness similar to anatomical reduction.
Humans ; Femur Head Necrosis ; Retrospective Studies ; Femoral Neck Fractures/surgery* ; Femur Neck ; Plastic Surgery Procedures

Objective:To retrospect the blood transfusion status of patients with extensive burns in multiple centers and analyze its influencing factors.Methods:A retrospective case series study was conducted. Clinical data of 455 patients with extensive burns who met the inclusion criteria and were admitted to the burn centers of 3 hospitals from January 2016 to June 2022 were collected, including 202 patients from the First Affiliated Hospital of Nanchang University, 179 patients from the Second Affiliated Hospital of Zhejiang University School of Medicine, and 74 patients from the First Affiliated Hospital of Anhui Medical University. The following data were collected from patients during their hospitalization, including infusion of red blood cells, plasma, and platelets during hospitalization; age, gender, body mass index, combined underlying diseases, cause of injury, time of admission after injury, type of admission, total burn area, full-thickness burn area, combination of inhalation injury, combination of other trauma, and combination of pulmonary edema; the blood lactic acid, serum creatinine, total bilirubin, and albumin values within 24 h of admission; combination of bloodstream, wound, lung, and urinary tract infection, and combination of sepsis; the number of escharectomy or tangential excision and skin grafting surgery (hereinafter referred to as surgery) and total surgical blood loss volume; occurrence of hemoglobin<70 g/L, admission to intensive care unit (ICU), conduction of mechanical ventilation and continuous renal replacement therapy (CRRT), length of hospital stay, and prognosis were recorded. In 602 surgeries of patients within 14 days after injury, data including area of escharectomy or tangential excision and skin graft harvesting, duration of operation, and surgical blood loss volume per surgery, operation site, and use of tourniquet and wound graft were collected. Data were statistically analyzed with Mann-Whitney U test, Kruskal-Wallis H test, and Spearman correlation analysis. Combined with the results of single factor analysis and clinical significance, multiple linear regression analysis was performed to screen the independent influencing factors of red blood cell infusion volume and plasma infusion volume, as well as blood loss volume per surgery. Results:During the whole hospitalization period, 437 (96.0%) patients received blood transfusion therapy, including 435 (95.6%) patients, 410 (90.1%) patients, and 73 (16.0%) patients who received transfusion of plasma, red blood cells, and platelets, respectively. The patients were mainly male, aged 18 to 92 years. There were statistically significant differences in the plasma infusion volume among patients with different combination of underlying disease, combination of inhalation injury, combination of other trauma, combination of pulmonary edema, combination of bloodstream infection, combination of wound infection, combination of lung infection, combination of urinary tract infection, combination of sepsis, occurrence of hemoglobin value <70 g/L, admission to ICU, conduction of mechanical ventilation, and conduction of CRRT (with Z values of -2.06, -4.67, -2.11, -6.13, -9.56, -4.93, -8.08, -4.78, -9.12, -6.55, -9.37, -11.46, and -7.17, respectively, P<0.05). The total burn area, full-thickness burn area, blood lactic acid value within 24 h of admission, serum creatinine value within 24 h of admission, albumin value within 24 h of admission, number of surgeries, and total surgical blood loss volume were correlated with the plasma infusion volume of patients (with r values of 0.39, 0.51, 0.14, 0.28, -0.13, 0.47, and 0.56, respectively, P<0.05).There were statistically significant differences in the red blood cell infusion volume among patients with different gender, combination of inhalation injury, combination of other trauma, combination of pulmonary edema, combination of bloodstream infection, combination of wound infection, combination of lung infection, combination of urinary tract infection, combination of sepsis, occurrence of hemoglobin value <70 g/L, admission to ICU, conduction of mechanical ventilation, and conduction of CRRT (with Z values of -2.00, -4.34, -3.10, -4.22, -8.24, -7.66, -8.62, -4.75, -7.42, -9.36, -6.12, and -8.31, -6.64, respectively, P<0.05). The age, total burn area, full-thickness burn area, blood lactic acid value within 24 h of admission, serum creatinine value within 24 h of admission, total bilirubin value within 24 h of admission, number of surgeries, and total surgical blood loss volume were correlated with the red blood cell infusion volume of patients (with r values of 0.12, 0.22, 0.49, 0.09, 0.18, 0.13, -0.15, 0.69, and 0.77, respectively, P<0.05). Combined underlying diseases, full-thickness burn area, combined pulmonary edema, serum creatinine value within 24 h of admission, combined sepsis, conduction of CRRT, number of surgeries, and total surgical blood loss volume were the independent influencing factors for plasma infusion volume during hospitalization in patients with extensive burns (with standardized regression coefficients of 0.09, 0.16, 0.12, 0.07, 0.11, 0.15, 0.31, and 0.26, respectively, P<0.05). Female, full-thickness burn area, serum creatinine value within 24 h of admission, combined sepsis, occurrence of hemoglobin value <70 g/L, conduction of CRRT, and total surgical blood loss volume were the independent influencing factors for red blood cell infusion volume during hospitalization in patients with extensive burns (with standardized regression coefficients of 0.10, 0.12, 0.10, 0.11, 0.05, 0.19, and 0.54, respectively, P<0.05). There were statistically significant differences in blood loss volume per surgery of patients with different surgical site and wound graft (with Z values of -2.54 and -2.27, respectively, P<0.05). The area of escharectomy or tangential excision and skin graft harvesting and duration of operation were correlated with the blood loss volume per surgery of patients (with r values of 0.40 and 0.21, respectively, P<0.05). The area of escharectomy or tangential excision and skin graft harvesting, duration of operation, and active wound grafts were the independent influencing factors for blood loss volume per surgery of patients with extensive burns (with standardized regression coefficients of 0.41, 0.16, and 0.12, respectively, P<0.05). Conclusions:The major factors influencing blood transfusion status in patients with extensive burns are female, combined underlying diseases, full-thickness burn area, serum creatinine value within 24 h of admission, combined pulmonary edema, occurrence of hemoglobin value <70 g/L, combined sepsis, conduction of CRRT, number of surgery, and total surgical blood loss volume. In addition, the area of escharectomy or tangential excision and skin graft harvesting, duration of operation, and active wound grafts indirectly affect the patient's blood transfusion status by affecting the blood loss volume per surgery.

Objective:The real-world clinical data of patients with newly diagnosed ovarian cancer (including fallopian tube cancer and primary peritoneal cancer) who received first-line maintenance therapy with poly adenosine diphosphate ribose polymerase inhibitor (PARPi) were retrospectively analyzed, and the prognostic factors were preliminarily explored.Methods:(1) The clinicopathological data and follow-up data of ovarian cancer patients treated with PARPi first-line maintenance therapy from August 2018 (PARPi was launched in China) to December 31, 2021 in Sichuan Cancer Hospital were collected (real-world clinical data). (2) According to the different types of PARPi, real-world clinical data were divided into olaparib group and niraparib group, which were respectively compared with the inclusion and exclusion criteria of representative domestic and foreign phase Ⅲ randomized controlled trials (RCT), including olaparib as first-line maintenance therapy for advanced ovarian cancer patients with BRCA1/2 gene mutation (SOLO-1 study), niraparib as first-line maintenance therapy (PRIMA study), and niraparib as first-line maintenance therapy for Chinese advanced ovarian cancer patients (PRIME study). (3) The prognosis of the two groups and the prognostic factors were analyzed.Results:(1) A total of 83 patients were included in this study, with a median age of 51 years (47-57 years), including 75 cases of ovarian cancer, 5 cases of fallopian tube cancer, and 3 cases of primary peritoneal cancer; 5 cases of stage Ⅰ, 9 cases of stage Ⅱ, 55 cases of stage Ⅲ, 12 cases of stage Ⅳ, and 2 cases of unknown stage; neoadjuvant chemotherapy (NACT) was performed in 40 cases and non-NACT in 43 cases; 62 cases had no visible residual lesion after surgery (R0), 9 cases had residual disease lesions <1 cm (R1), 8 cases had residual disease lesions ≥1 cm (R2), and 4 cases with unknown postoperative residual disease. Thirty-two cases had PARPi treatment interruption, 40 cases had PARPi reduction, and 1 case terminated treatment due to acute leukemia. Of the 83 patients, 35 were in the olaparib group and 48 were in the niraparib group. The proportion of patients with high-grade serous carcinoma (100% and 75%, respectively) and the proportion of BRCA mutant patients (91% and 10%, respectively) in the olaparib group were higher than those in the niraparib group (all P<0.01). (2) Compared with the inclusion and exclusion criteria of the SOLO-1 study, the olaparib group had only 60% (21/35) coincidence rate; compared with the inclusion and exclusion criteria of PRIMA and PRIME studies, the coincidence rates of niraparib group were only 31% (15/48) and 69% (33/48). The most common reasons for non-compliance were number of chemotherapy courses, histopathological type, and surgical pathological stage. (3) Of the 83 cases received first-line maintenance therapy with PARPi, the median follow-up was 15.9 months (11.3-22.9 months), the median progression-free survival (PFS) was 29.7 months (95% CI: 25.9-33.6 months), and the median overall survival was 49.8 months (95% CI: 47.4-52.2 months). Univariate analysis showed that unilateral or bilateral ovarian cancer, efficacy after platinum-containing chemotherapy, presence or absence of measurable lesions at the end of chemotherapy, and total number of chemotherapy courses were significantly associated with PFS (all P<0.05). Multivariate analysis showed that unilateral or bilateral ovarian cancer, total number of chemotherapy courses, and efficacy after platinum-containing chemotherapy were independent factors affecting PFS in stage Ⅱ-Ⅳ patients with PARPi first-line maintenance therapy (all P<0.05). Conclusions:Unilateral ovarian cancer, the total number of chemotherapy courses no more than 9, and achieving complete response after platinum-containing chemotherapy before maintenance therapy are independent influencing factors of PFS benefit in patients with PARPi first-line maintenance therapy. Due to the large differences between the patients in real clinical practice and the research subjects of phase Ⅲ RCT, the results of representative retrospective studies still have important clinical reference significance.

Objective:To explore the risk factors in leukemia transformation (LT) in those with myelodysplastic syndromes (MDS) .Methods:From January 2012 to December 2020,data on 320 patients with newly diagnosed primary MDS were gathered from the MDS center. The clinical features and molecular characteristics are explored. Additionally, a retrospective analysis of risk factors for the development of acute leukemia from MDS was done.Results:The median follow-up was13.6 (0.4-107.3) months. 23.4% (75/320) of the MDS patients had LT group. Significant differences between the LT group and non-LT group can be seen in age ( P<0.001) , bone marrow blast percentage ( P<0.001) , bone marrow fibrosis ( P=0.046) , WHO classification ( P<0.001) , IPSS-R ( P<0.001) and IPSS-R karyotype group ( P=0.001) . The median number of mutation of LT group was 1 (1, 3) , that in non-LT group was 1 (0, 2) ,which had a statistical difference ( P=0.003) .At the time of the initial diagnosis of MDS, the LT group had higher rates of the TP53 mutation ( P=0.034) , DNMT3A mutation ( P=0.026) , NRAS mutation ( P=0.027) and NPM1 mutation ( P=0.017) . Compared with the mutations at first diagnosis and LT of six patients, the number of mutations increased and the variant allele frequencies (VAF) increased significantly in LT patients. Higher bone marrow blast percentage (Refer to <5% , 5% -10% : HR=4.587, 95% CI 2.214 to 9.504, P<0.001, >10% : HR=9.352, 95% CI 4.049 to 21.600, P<0.001) , IPSS-R cytogenetic risk groups ( HR=2.603, 95% CI 1.229-5.511, P=0.012) , DNMT3A mutation ( HR=4.507, 95% CI 1.889-10.753, P=0.001) , and NPM1 mutation ( HR=3.341, 95% CI 1.164-9.591, P=0.025) were all independently associated with LT in MDS patients, according to results of multivariate Cox regression. Conclusion:Bone marrow blast percentage, IPSS-R cytogenetic risk groups, DNMT3A mutation, and NPM1 mutation are independent risk factors in LT for MDS patients.

Objective:Diagnostic value assessment of sternal bone marrow cell morphology in patients with acquired hypocellular bone marrow failure syndromes (BMFS) characterized by normal cytogenetics.Methods:A total of 194 eligible patients with an acquired hypocellular BMFS pre-sternum diagnosis in Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College from June 2014 to January 2019 were reviewed. Sternal bone marrow evaluation was performed, and a post-sternum diagnosis was made. Clinical characteristics and overall survival (OS) were then compared among patients with different post-sternum diagnosis. Binary logistic regression was used to develop a predictive scoring system.Results:In 152 patients with pre-sternum AA diagnosis, 29 patients with a pre-sternum idiopathic cytopenia of undetermined significance (ICUS) diagnosis, and 13 patients with a pre-sternum clonal cytopenia of undetermined significance (CCUS) diagnosis, sternal bone marrow evaluation resulted in a change of diagnosis to hypocellular myelodysplastic syndrome (hypo-MDS) in 42.8% (65/152) , 24.1% (7/29) , and 30.8% (4/13) , respectively. Patients with a post-sternum hypo-MDS diagnosis showed a significant difference in OS compared with patients with a post-sternum AA diagnosis ( P=0.005) . Patients with ICUS/CCUS showed no difference in OS compared with AA and hypo-MDS ( P=0.095 and P=0.480, respectively) . A 4-item predictive scoring system to identify hypocellular BMFS patients that need sternal bone marrow evaluation was developed, including age > 60 years old ( OR=6.647, 95% CI 1.954-22.611, P=0.002, 2 points) , neutrophil alkaline phosphatase score ≤ 160 ( OR=2.654, 95% CI 1.214-5.804, P=0.014, 1 point) , abnormal erythroid markers evaluated by flow cytometry on iliac bone marrow ( OR=6.200, 95% CI 1.165-32.988, P=0.032, 2 points) , and DAT (DNMT3A, ASXL1, TET2) genes mutation ( OR=4.809, 95% CI 1.587-14.572, P=0.005, 1 point) . The Akaike information criterin (AIC) was 186.1. Conclusion:Patients with a pre-sternum acquired hypocellular BMFS diagnosis characterized by normal cytogenetics may not reach accurate diagnostic categorization without sternal bone marrow cell morphology evaluation, which could be considered a diagnostic tool for this patient population. A predictive scoring system was developed, and when the total score is ≥ 2 points, sternal bone marrow evaluation should be performed for accurate diagnostic categorization that is critical to optimal patient care.

Objective:To evaluate the prognostic value of MIPSS70-plus in Chinese patients with primary myelofibrosis (PMF) .Methods:A total of 113 Chinese patients with PMF were retrospectively analyzed. The Kaplan-Meier method, Log-rank test, and Cox proportional hazard regression model were performed to evaluate the prognostic factors. The likelihood ratio test was used to evaluate the predictive power between MIPSS70-plus and DIPSS systems.Results:The median age of the Chinese patients was 55 (range: 20-70) years, including 71 males and 42 females. According to the standard of MIPSS70-plus system, 99 patients (79.6% ) had a favorable karyotype and 23 patients (20.4% ) had an unfavorable karyotype. JAK2V617F in 55.8% ( n=63) , CALR exon9 in 17.7% (including 15 CALR type 1 and 5 CALR type 2, n=20) , MPLW515 in 4.4% ( n=5) , and triple negative (no detectable JAK2, MPL, and CALR mutations) in 22.1% of patients in our cohort were found by target-specific next-generation sequencing approach. At least one high-molecular risk mutations were presented in 45.1% ( n=51) of patients, with ASXL1 in 38.9% ( n=44) , SRSF2 in 7.1% ( n=8) , IDH1/2 in 4.4% ( n=5) , and EZH2 in 3.5% ( n=4) of patients. A total of 28 patients (26.7% ) were in low risk, 20 (19.0% ) in intermediate risk, 41 (39.0% ) in high risk, and 16 (15.3% ) in very-high risk categories, which were delineated for the MIPSS70-plus model. A 2-year OS was 100% in low risk, 89.7% (95% CI 76.2% -100.0% ) in intermediate risk, 64.8% (95% CI 47.0% -82.6% ) in high risk, and 35.0% (95% CI 10.3% -59.7% ) in very-high risk categories, which had a significant difference ( P<0.001) . A significantly higher predictive power for survival of the MIPSS70-plus group was observed compared with the DIPSS group ( P=0.001, -2 log-likelihood ratios of 86.355 vs 95.990 for the MIPSS70-plus and DIPSS systems, respectively) . Conclusion:The MIPSS70-plus had significantly higher predictive power than the DIPSS.

Objective:To explore the outcome of cyclosporine A (CsA) combined with danazol with or without thalidomide regimen for myelodysplastic syndrome (MDS) with low-percentage bone marrow blasts and predictive factors for treatment response.Methods:Data of 115 subjects who were newly diagnosed with primary MDS with low-percentage bone marrow blasts and were treated with CsA combined with danazol with or without thalidomide from December 2011 to December 2019 in our center were collected. Their clinical features, efficacy, and predictive factors of efficacy were retrospectively analyzed. A model for predicting this response was developed.Results:A total of 55 subjects responded (47.8%) , including 11 complete responses and 44 hematologic improvements. Fifty-two patients (52/105, 49.5%) achieved erythrocyte response; 35 (35/86, 40.7%) , platelet response; and 14 (14/40, 35%) , neutrophil response. Of 29 subjects (24.1%) , 7 who were red blood cell (RBC) transfusion-dependent became independent of transfusion. The median response duration was 20 months (range, 3-84 months) . In the univariate analysis, patients <0 years had a higher response rate than those ≥60 years (52.5% vs 22.2%, P=0.018) . Contrarily, the response rate was substantially decreased in patients with RBC transfusion dependence compared with those without RBC transfusion dependence (24.1% vs 55.8%, P=0.003) , as well as in patients with the mutated U2AF1 compared with those with the wild-type U2AF1 (26.1% vs 53.2%, P=0.020) . In multivariable analyses, age <0 years ( OR=4.302, 95% CI 1.245-14.820, P=0.021) , RBC transfusion dependence ( OR=3.774, 95% CI 1.400-10.177, P=0.009) , and U2AF1 mutation ( OR=3.414, 95% CI 1.168-9.978, P=0.025) were significantly correlated with response. Variables that independently predicted the response were combined to generate the predictive model. According to the model, the overall response rates of patients with 0, 1, 2, and 3 risk factors were 65%, 30%-35%, 10%-15%, and 3%, respectively. Conclusion:CsA combined with danazol with or without thalidomide regimen could improve cytopenia symptoms in patients with MDS with low-percentage bone marrow blasts. At age <60 years, no transfusion dependence of RBC and wild-type U2AF1 mutation is a favorable prognostic factor.

Objective:To explore the relationship between symptom burden and hematologic responses after treatment with interferon and/or hydroxyurea in patients with polycythemia vera (PV) .Methods:Hematologic responses after continuous treatment with interferon and/or hydroxyurea for six months were evaluated in 190 patients with PV using the Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score (MPN-10 score) . In all patients, the PV diagnosis was based on the 2016 World Health Organization diagnostic definitions.Results:The study cohort comprised 93 (48.9% ) male and 97 (51.1% ) female patients. The median age at the time of MPN-10 assessment was 60 (32-82) years. The median MPN-10 score of the entire cohort was 9 (range, 0-67) . The median MPN-10 score of patients treated with interferon plus hydroxyurea ( n=27) was 11 (0-67) , which was significantly higher than those of patients treated with interferon only ( n=64) (6[0-56], P=0.019) or hydroxyurea only ( n=99) (9[0-64], P=0.047) , whereas the median MPN-10 score was not significantly different between those treated with interferon only and hydroxyurea only ( P=0.421) . The rate of severe symptom burden (i.e., any single symptom burden score ≥ 7 and/or total score ≥ 44) was 28.9% (55/190) in the entire cohort, whereas the rate of severe symptom burden was not significantly different among the interferon only (23.4% ) , hydroxyurea only (29.3% ) , and interferon plus hydroxyurea (40.7% ) groups ( P>0.05 for all two-group comparisons) . When evaluating MPN-10 score, 37.4% (71/190) of the patients achieved complete hematologic remission (CHR) . Only 28.9% (55/190) patients had adequate disease control, defined as CHR without severe symptom burden. Reasons for inadequate disease control were evaluating blood counts alone, severe symptom burden alone, and evaluating blood counts accompanied with severe symptom burden in 42.1% (80/190) , 8.4% (16/190) , and 20.5% (39/190) of the patients, respectively. Compared to the patients with a platelet count ≤ 400×10 9/L, those with a platelet count > 400×10 9/L had a significantly higher rate of severe symptom burden (40.8% [20/49] vs 24.8% [35/141], P=0.044) and a higher median MPN-10 score (14[0-67] vs 7[0-56], P=0.038) . Platelet count > 400×10 9/L was associated with an increased risk of severe symptom burden (hazard ratio, 2.089; 95% confidence interval, 1.052-4.147, P=0.035) . Conclusions:Symptoms related to disease after treatment with interferon and/or hydroxyurea were rather universal in patients with PV. Some patients still experienced severe symptom burden despite achieving CHR. Platelet count > 400×10 9/L was associated with an increased risk of severe symptom burden in patients with PV treated with interferon and/or hydroxyurea.

Objective:To explore predictors of overall survival (OS) in Chinese patients with polycythemia vera (PV) .Methods:A total of 906 consecutive newly diagnosed patients with PV seen at the Blood Diseases Hospital, Chinese Academy of Medical Sciences, from June 2007 to February 2020 were included, and their data were collected. PV was diagnosed according to 2016 World Health Organization (WHO) diagnostic definitions. OS and prognostic factors were retrospectively analyzed.Results:Among the 906 patients, 439 were male (48.5%) and 467 were female (51.5%) . The median age was 57 years (range: 18-91 years) . 31.6% (276/874) of the patients had a thrombosis history at diagnosis, and 4.6% (25/541) of the patients had abnormal cytogenetics. The median follow-up was 54 months (95% confidence interval [ CI] 8-130 months) . The 5- and 10-year cumulative deaths were 5.8% (95% CI 4.8%-6.7%) and 11.1% (95% CI 9.3%-12.9%) , respectively. Univariate analysis showed that age ≥60 years, thrombosis history, white blood cells (WBC) ≥15×10 9/L, platelet (PLT) ≥450×10 9/L, and platelet distribution width (PDW) ≥15 fl significantly correlated with worse OS, and palpable spleen correlated with better OS. Multivariate analysis showed that age ≥60 years ( HR=4.3, 95% CI 2.1-9.2, P<0.001) and PDW ≥15 fl ( HR=2.1, 95% CI 1.1-4.0, P=0.023) were independent prognostic factors for worse OS. The 5-year cumulative death for patients with PDW ≥15 fl or PDW<15 fl was 8.6% (95% CI 5.9%-11.3%) or 4.4% (95% CI 3.4%-5.4%) , respectively. The 5-year cumulative death for patients defined as low-, intermediate-, and high-risk patients by international working group score system for PV (IWG-PV) were 0.8% (95 CI 0.2%-1.4%) , 4.0% (95% CI 2.7%-5.3%) , and 12% (95% CI 9.6%-14.4%) , respectively, with a significant difference among the three cohorts ( P<0.05) . PDW ≥ 15 fl significantly affected OS for intermediate- and high-risk patients ( HR=2.3, 95% CI 1.2-4.2, P=0.009) defined by IWG-PV score system, but not for low-risk patients ( HR=3.1, 95% CI 0.2-52.0, P=0.405) . Conclusions:Age ≥60 years and PDW ≥15 fl were independent prognostic factors for worse OS in PV. IWG-PV score system effectively predicted OS for Chinese patients with PV.