Myelodysplasias (MDS)

Refinement of the international prognostic scoring system (IPSS) by including LDH as an additional prognostic variable to improve risk assessment in patients with primary myelodysplastic syndromes (MDS)

Article metrics


The international prognostic scoring system (IPSS) is considered the gold standard for risk assessment in primary myelodysplastic syndromes (MDS). This score includes several prognostic factors except serum lactate dehydrogenase (LDH). We evaluated the prognostic power of LDH as an additional variable in IPSS-based risk assessment. For this purpose, a total of 892 patients with primary MDS registered by the Austrian–German cooperative MDS study group was analyzed retrospectively. Multivariate analysis confirmed the value of established parameters such as medullary blasts, karyotype and peripheral cell counts and showed that elevated LDH was associated with decreased overall survival (P<0.00005) and increased risk of AML development (P<0.00005), independent of the system used to classify MDS (FAB or WHO). Moreover, elevated LDH was found to be a significant predictor of poor survival within each IPSS risk group and within each FAB group except RAEB-T. To exploit these results for refined prognostication, each IPSS risk group was split into two separate categories (A=normal LDH vs B=elevated LDH). Using this LDH-assisted approach, it was possible to identify MDS patients with unfavorable prognosis within the low and intermediate IPSS risk groups. We propose that the IPSS+LDH score should improve clinical decision-making and facilitate proper risk stratification in clinical trials.


The international prognostic scoring system (IPSS),1 which was introduced in 1997, has become the ‘gold standard’ for risk assessment in patients with primary myelodysplastic syndromes (MDS). This score has improved risk stratification in clinical trials and is also used for decision-making in clinical practice. When the IPSS was developed, databases from seven working groups, each of which had already proposed a scoring system for assessing survival and evolution to acute myeloid leukemia (AML) in previous analyses2, 3, 4, 5, 6, 7, 8 were employed.1 The IPSS includes three cytogenetic risk categories together with ‘classic’ prognostic parameters, namely medullary blast count and number of cytopenias. More recently, a working party of the World Health Organization (WHO) proposed a new MDS classification.9, 10 Following this WHO proposal, chronic myelomonocytic leukemia (CMML) and refractory anemia with excess blasts in transformation (RAEB-T) were eliminated from the group of MDS. So far, it is unknown, however, whether the IPSS is applicable to patients diagnosed according to the new WHO classification in the same way as it is to patients diagnosed according to the French–American–British (FAB) classification.2

Although several studies have confirmed the general value of the IPSS, it is still difficult to predict survival and AML development in individual patients with MDS. Especially the prognostication of patients with dysplastic CMML remains difficult.11 Therefore, it seems desirable to search for additional prognostic variables which might be considered for inclusion into the IPSS to improve its prognostic power. One of these variables is the serum level of lactate dehydrogenase (LDH) activity. In fact, a number of studies have shown that elevated LDH is associated with poor prognosis in MDS.6, 12, 13, 14, 15, 16, 17, 18 In order to validate and possibly improve prognostication in MDS, a working party of seven Austrian and German groups who maintain MDS registries has been established recently. Using the database of this working group, cytogenetic, hematological and morphological data of a larger number of patients with MDS diagnosed in these centers, were collected and then analyzed centrally. The specific aims of the current study were (i) to define the value of elevated LDH as an additional prognostic variable in IPSS-based risk assessment, (ii) to refine the IPSS by including the LDH, and (iii) to assess whether the new LDH-adjusted IPSS would yield the same results when excluding patients with CMML and RAEB-T, according to the new WHO proposal to classify MDS.

Patients and methods


A total of 1247 patients with MDS, who were diagnosed between 1975 and 2002, were included in a joint database of the Austrian–German working group and examined in a retrospective manner. Of these patients, 94 (7.5%) were classified as therapy-related secondary MDS, attributed to previous chemotherapy and/or radiotherapy, and therefore were excluded from the analysis. A total of 261 patients (21%) who received intensive chemotherapy or stem cell transplantation for treatment of primary MDS were analyzed separately. The remaining 892 patients with primary MDS, including patients treated with hematopoietic growth factors, represented the primary sample set for risk analysis.

All source data were collected in local registries in seven regional centers of the Austrian–German MDS working group. Source data included case history (including mutagenic events) and clinical features. As a prerequisite for inclusion in the study, cytogenetic data as well as other data required for applying the different scoring systems, had to be recorded. Cytopenias were defined according to the IPSS (hemoglobin level <10 g/dl, absolute neutrophil count (ANC) <1800/μl, platelets <100 000/μl). All variables (cell counts, LDH, medullary blasts, FAB type, cytogenetic findings) were assessed at the time of diagnosis. Minimal follow-up information included survival, as measured from the time of diagnosis of MDS, confirmed by bone marrow cytology. It also included the time from MDS diagnosis to AML evolution. The median time of follow-up was 25 months. After collection, the complete data sets were forwarded to the central statistician (PM). Clinical and cytogenetic data and categorization of prognostic parameters are summarized in Table 1.

Table 1 Patient characteristics and their impact on survival/AML evolution (univariate analysis)

Diagnosis of MDS

The diagnosis and category of MDS were established according to criteria provided by the FAB cooperative study group.2 The medullary blast count was determined on the basis of at least 300 nucleated cells analyzed in bone marrow smears. A diagnosis of MDS required that cytopenia had to be present for at least 3 months. All MDS types including both variants of CMML, that is, the dysplastic and proliferative subtype, were considered in the analysis. In order to allow direct comparison with the original IPSS data set,1 the analysis was performed with and without cases of myeloproliferative CMML. Additionally, we analyzed the data of 660 patients, after elimination of RAEB-T and CMML cases, following the WHO proposal to classify MDS. For cross validation, each center submitted representative bone marrow slides for each MDS type (at least five cases for each FAB category), which were reviewed by the central morphologist (CA). This cross-validation was adapted from the methods of the IPSS group and yielded no discrepancies regarding morphological diagnoses according to FAB types. AML evolution was defined by an increase of the medullary blast count to 30% or more.


Cytogenetic analysis was performed in local centers. The results were reported to the central registry and reviewed by a central genetic board (DH, BH, CS). A minimum of 10 metaphases were required for analysis and reporting. Cases with less than 10 metaphases were excluded from the analysis. The median number of metaphases was 22. Cytogenetic findings were documented according to the International System for Human Genetic Nomenclature. They were classified following the IPSS proposal (low risk: 5q−, 20q, −Y, and normal karyotype; high risk: aberrations of chromosome 7 and/or complex karyotypes, that is, 3 abnormal chromosomes; intermediate risk: all other findings). For a more detailed risk analysis, we separated karyotypes into 10 different categories, namely: normal; 5q−; −Y; 20q−; +8; chromosome 7 abnormalities; miscellaneous double anomalies; 11q23 anomalies; complex karyotypes.

Analysis of LDH activity

The serum levels of LDH activity were determined by a commercial assay (Boehringer-Mannheim, Germany). The LDH enzyme activity was quantified by measuring the knock-down of NADH in the catalytic enzyme reaction

by photometry. The decrease in NADH in this reaction is proportional to the LDH concentration in the sample. The LDH activity was given in units per liter (U/l). The reference standard and cutoffs determining the upper normal value of LDH differed slightly from center to center. In the present study, a cutoff of 240 U/l was selected for discriminating normal from elevated LDH levels. The median LDH level in all MDS patients was 201 U/l. In 31% of all patients, the values were above 240 U/l (=upper limit of normal).

Statistical analysis

All clinical, hematological and cytogenetic parameters were analyzed by the central statistician (PM) using the log-rank test. Categorization and definition of cutpoints were based on clinical as well as statistical considerations. P-values below 0.05 were considered statistically significant. Multivariate analyses were performed using the proportional hazard regression analysis (multiple Cox regression). The final weighting of the prognostic impact of variables was based on clinical and practical considerations (rounding up cutpoints and figures) as well as on statistical grounds.


Patients and classification of MDS

Source data from 892 patients with primary MDS were analyzed for risk assessment in this study. According to the FAB classification, 338 patients were diagnosed as refractory anemia (RA) (37%), 152 as refractory anemia with ringed sideroblasts (RARS) (17%), 170 as refractory anemia with excess blasts (RAEB) (19%), 82 as RAEB-T (9%), and 150 as CMML (17%). The median age at diagnosis was 68 years (range: 16–95), and the male/female ratio was 1.3 : 1. A total of 478 patients died within the observation period (54%). Evolution to AML was observed in 204 patients (23%). For analysis of cumulative AML development (Kaplan–Meier estimates), 47 patients had to be censored because the time of AML-transformation was not recorded.

Identification and evaluation of risk factors by univariate analysis

Univariate analysis identified the following determinants as adverse prognostic factors for survival: advanced FAB type (RAEB/RAEB-T), elevated medullary blast count (>5%), low hemoglobin concentration (<9 g/dl), age >70 years, male gender, elevated serum LDH activity, low platelet count (<100 000/μl), ANC <1800/μl, more than one cytopenia, and a complex karyotype. With regard to AML development, adverse prognostic factors were advanced FAB subtype, elevated medullary blast count, low hemoglobin, elevated LDH activity, ANC <1800/μl, more than one cytopenia, and a complex karyotype (Table 1).

The influence of the cytogenetic risk category defined by the IPSS on survival and risk of AML transformation is clearly demonstrated: Patients in the cytogenetic low-risk group as defined by the IPSS had a median survival of 62 months, as compared to 31 months for patients in the intermediate-risk group, and 11 months for patients in the high-risk group. Those with 5q− as sole anomaly had the longest median survival, namely 84 months. Trisomy 8 was found to predict a poor prognosis, with a median survival of 24 months.

Within the large group of diverse and sporadic abnormalities, there were no particular aberrations indicating a good or bad prognosis. The worst median survival (9 months) was seen in patients with complex karyotypes. These patients also exhibited the highest risk to transform to AML. The latter was also found to be true for patients with abnormalities of chromosome 7. As shown in Figure 1a and b, elevated LDH activity was also associated with poor prognosis regarding survival and AML evolution. While age at diagnosis had an influence on survival, there was no association between age and risk of AML development.

Figure 1

(a) Kaplan–Meier estimates of survival according to LDH activity. (b) Kaplan–Meier estimates of AML evolution according to LDH activity.

Assessment of risk factors by multivariate analysis

For multivariate analysis, all parameters that showed prognostic significance by univariate testing, were included (Table 2). Survival was independently influenced by the cytogenetic risk category, medullary blast count, LDH activity, age, hemoglobin level, gender, and platelet count, whereas ANC did not enter the regression model.

Table 2 Multivariate analysis of different prognostic parameters for survival and AML evolution. (patients with myeloproliferative CMML excluded)

We also performed multivariate analysis for the different FAB types separately. For patients with RA, the cytogenetic category, gender, LDH activity, and platelet count entered the regression model. In RARS, gender, hemoglobin level, and LDH activity were shown to be independent prognostic markers. For patients with RAEB and RAEB-T, the cytogenetic category and LDH activity proved to be important prognostic variables. In CMML, the cytogenetic category was not found to be predictive, perhaps due to the relatively low proportion of abnormal karyotypes, whereas the hemoglobin level and the LDH activity were independent prognostic parameters. Regarding transformation to AML, multivariate analysis showed that medullary blast count, cytogenetic risk group, gender, LDH activity and hemoglobin level were the most important prognostic factors.

Application of the IPSS and the Düsseldorf score

When applying the IPSS, patients with CMML and a WBC>12 000/μl (myeloproliferative subtype, n=76) were excluded from (this part of) the analysis, since the IPSS was developed without such patients. We confirmed that the IPSS separates four risk groups with different prognosis in terms of survival and AML development (Figure 2a and b, Table 1). The low-risk group had a median survival of 88 months and a cumulative risk of AML of 8% after 5 years, whereas patients in the IPSS high-risk group had a median survival of 13 months and a cumulative risk of AML of 60% after 5 years. The Düsseldorf score, which does not include cytogenetics, was particularly effective in identifying a group of patients (19%) with favorable prognosis, whose median survival was 108 months. At the other end of the spectrum, the Düsseldorf high-risk group (median survival of 13 months) included a larger number of patients (22%) compared with the high-risk group of the IPSS.

Figure 2

(a) Kaplan–Meier estimates of survival according to IPSS risk groups (low risk=A, intermediate risk I=B, intermediate risk II=C, high risk=D). (b) Kaplan–Meier estimates of AML evolution according to IPSS risk groups.

In a further step, we applied the IPSS and the Düsseldorf score to different FAB types separately. The IPSS was able to subdivide RARS and RAEB, respectively, into two risk groups with significantly different prognosis. The Düsseldorf Score was able to subdivide RA, RARS, RAEB and CMML, respectively, into two or three risk groups (data not shown).

Proposed adjustment of the IPSS by including the LDH level to split subgroups

When we combined parameters that had shown prognostic impact in our multivariate analysis, we found that all four IPSS risk groups can be subdivided into two prognostically different subgroups (A and B), according to normal (A) or an elevated (B) LDH level (Table 3a). Figure 3a–d shows the respective Kaplan–Meier plots for each IPSS risk group, divided into A and B groups. The median survival of the IPSS low-risk groups was 107 months (low-A) and 63 months (low-B), respectively (P<0.0011), the median survival of the intermediate I risk groups was 66 months (int-I-A) and 36 months (int-I-B), respectively (P=0.0006), the median survival of the intermediate II risk groups were 26 months (int-II-A) and 16 months (int-II-B), respectively (P=0.01), and the median survival of the high-risk groups was 16 months (high-A) and 11 months (high-B), respectively (P=0.01). Besides that, the IPSS+LDH allowed us to separate risk groups within all FAB types except RAEB-T. In addition, we tested the IPSS+LDH in older and younger patients. The IPSS+LDH was found to separate risk groups in patients <60 years as well as in patients 60 years. Median survival in the high and intermediate II risk groups was similar for younger and older patients. A survival advantage for younger patients was only seen in the low and intermediate I risk groups.

Table 3 Prognosis of risk groups according to IPSS and IPSS+LDH (a) for all FAB Types and (b) only WHO Types included, without CMML and RAEB-T
Figure 3

(a) Kaplan–Meier estimates of survival in the IPSS low-risk group according to LDH activity (A=normal, B=elevated). (b) Kaplan–Meier estimates of survival in the intermediate I risk group according to LDH activity (A=normal, B=elevated). (c) Kaplan–Meier estimates of survival in the IPSS intermediate II risk group according to LDH activity (A=normal, B=elevated). (d) Kaplan–Meier estimates of AML evolution in the IPSS high-risk group according to LDH activity (A=normal, B=elevated).

Application of the new IPSS+LDH scoring system to patients after elimination of RAEB-T and CMML (WHO proposal)

In an attempt to define the value of the new IPSS+LDH score system in light of the new WHO classification, we analyzed MDS patients after elimination of RAEB-T and CMML. In these 660 patients, the new IPSS+LDH was also found to be a most powerful scoring system with regard to predicting survival and AML development. In particular, the prognosis of the resulting risk groups differed significantly (Table 3b) with similar outcomes as compared to patients classified according to the FAB system.

Prognostic parameters in patients receiving intensive chemotherapy

Finally, we analyzed 261 patients with MDS who were treated with intensive chemotherapy with or without transplantation of hematopoietic stem cells during the course of their disease. All treatment regimens were based on Ara-C plus an Anthracycline with or without Etoposid. Although the IPSS was not created for predicting therapy outcome but for assessing the prognosis of untreated patients, it is widely used within different treatment studies, including chemotherapy and transplantation studies. The end point studied here was the overall survival. The median age in this group was 57 years and the median medullary blast count was 15%. An abnormal karyotype was found in 55% of the patients, while 29% were found to belong to the high-risk cytogenetic group as defined by the IPSS. According to the IPSS, 5% of the patients initially belonged to the low-risk group, 19% to the INT-1, 19% to the INT-II, and 57% to the high-risk group. On multivariate analysis, there were only three parameters that predicted the outcome of patients receiving chemotherapy: complex karyotype, platelet count <100 000/μl, and age >60 years. None of the other parameters were associated with prognosis. As a result, neither the IPSS nor the Düsseldorf score appeared to define risk groups in this patient population.


Although the IPSS is an accepted gold standard for risk assessment in MDS, several risk parameters are not included, and only a few confirmatory studies on MDS patient groups have been presented so far. Based on the data of 892 patients collected by the Austrian–German collaborative study group, which is similar in size compared to the IPSS database, we confirmed the value of the IPSS in risk assessement. Thus, based on multivariate analysis, we showed that medullary blast count is the most important parameter. Our data also confirm that cytogenetic categories are an important component of risk assessment in MDS. In addition, our data provide evidence that inclusion of LDH as a prognostic parameter results in a refinement of the IPSS. Thus, using this new IPSS+LDH, it should be possible to predict survival and AML-development more accurately in clinical practice.

Karyotyping is a well-recognized prognostic variable in risk assessment in MDS.1, 5, 7, 8, 12 In the current study, patients with a complex karyotype or aberrations of chromosome 7 had a very poor prognosis, whereas the prognosis was uniformly good when patients presented with 20q−, 5q−, −Y, or a normal karyotype. These data are in line with the IPSS1 as well as with the results obtained by Sole et al,19 Haase et al20 and Oguma et al.21 The very good prognosis of patients with 5q− as sole anomaly was only demonstrable in patients with a normal medullary blast count, a finding which was recently also reported by Giagounidis et al.13 In our multicenter study, patients with 5q− syndrome had a median survival of 107 months, as compared to 18 months in patients with a 5q− anomaly in the context of RAEB. Within the group of patients showing a broad spectrum of different aberrations involving one or two chromosomes, it was not possible to define particular prognostic subgroups. Patients in this heterogeneous cytogenetic group had an intermediate prognosis and a significantly better outcome than patients with three or more chromosomal abnormalities. Altogether, cytogenetic analysis is another important risk factor for predicting survival and AML development in MDS and should thus be performed whenever possible. However, our data show that the Düsseldorf score, a scoring system that does not depend on cytogenetics, is also capable of defining clinically relevant risk groups. For example, as compared with the IPSS, the Düsseldorf score separated a larger group of high-risk patients with very poor prognosis, and also defined a group of low-risk patients whose median survival was almost two years longer than that of low-risk patients in the IPSS. Interestingly, the low-risk group in the Düsseldorf score identified many patients with normal karyotype who had a good prognosis. We therefore suggest that the Düsseldorf score can be used if karyotyping results are not available. There was no strong correlation between the cytogenetic risk category and levels of LDH activity, which therefore provides additional prognostic information. Up to now it remains largely unknown why LDH is such a strong prognostic parameter in several hematological malignancies. To our knowledge, it is not even clear whether the degree of spontaneous tumor cell lysis is the major determinant, or whether increased expression of LDH in malignant hematopoietic cells contributes to elevated LDH activity in the serum. Both physical instability and metabolic changes may correlate with a particularly malignant phenotype.

Although several investigators have suggested that LDH activity is a useful prognostic parameter in MDS using different cutpoints,6, 14, 15, 16, 17, 18, 22, 23 it is for the first time that the prognostic significance of LDH was compared to that of cytogenetic findings in a large MDS database. As shown in Table 2, we found that LDH was nearly as powerful as a prognostic parameter when compared to the prognostic value of karyotyping. Furthermore, LDH activity had greater impact than cell counts on multiregression analysis. This was true for the entire group as well as for all FAB types evaluated separately, except for RAEB-T.

Since our results clearly pointed to the usefulness of LDH for risk assessment in MDS, we tried to harness the prognostic power of LDH to improve further the prognostic value of the IPSS by including LDH as an additional variable and make a good system even better.

Thus, when integrating LDH values as a new discriminating parameter, the new IPSS+LDH yielded convincing results. For example, the IPSS+LDH was capable of separating a new group of ‘very-low-risk’ patients, including patients with RA or RARS with a favorable karyotype and normal LDH, whereas an increased LDH was found to be associated with a worse prognosis and signified impending danger. In other words, patients with elevated LDH tended to have a worse prognosis than predicted by the IPSS. Especially in cases with less than 5% marrow blasts, elevated LDH identified poor-risk patients with great accuracy. Also in patients of the intermediate II and high-risk group, the addition of the LDH results in meaningful separations with differences in survival time of about 62 and 45%, respectively.

As another advantage, inclusion of LDH makes the scoring system less dependent on an exact evaluation of the medullary blast count. This may be clinically relevant, since the bone marrow smear is sometimes difficult to be evaluated in terms of blast cell counts, for example when the bone marrow aspirate is contaminated with peripheral blood cells or was taken from fibrotic marrow.

As far as blast counts are concerned, it should be mentioned that when the IPSS was developed, dysplastic CMML and RAEB-T were included, and a medullary blast count of more than 20% was chosen as the most powerful component of the scoring system. To our knowledge, the IPSS has never been re-evaluated in a large population of patients with MDS diagnosed according to the WHO classification.9, 10 In our study, the proportion of IPSS high-risk patients was diminished when RAEB-T patients were excluded. However, the median survival of the remaining IPSS high-risk patients remained almost unchanged. We also were able to confirm the good prognosis in patients with isolated 5q−, which is a separate category defined by the WHO. All in all, although we were not able to define the exact WHO category of MDS in all patients, we believe that the new IPSS+LDH can readily be applied to MDS patients diagnosed according to WHO criteria.

The IPSS as well as the Düsseldorf score failed to separate different risk groups in patients treated with intensive chemotherapy. This was attributable to the fact that only the karyotype remained predictive in this group of patients, whereas all other factors lost their prognostic impact. At present, it is unknown whether this failure is restricted to patients receiving intensive chemotherapy or extends to other therapeutic approaches that go beyond supportive care and hematopoietic growth factors. Since more and more patients with MDS receive experimental treatment, improved tools for risk assessment are clearly needed. Strictly speaking, any kind of investigational treatment may result in new prognostic parameters and scoring systems.

In summary, our data show that, for predicting the natural course of MDS, inclusion of LDH as a powerful prognostic parameter refines and improves the IPSS. This should be relevant for clinical decision-making and should facilitate proper risk stratification in clinical trials.


  1. 1

    Greenberg P, Cox C, Le Beau MM, Fenaux P, Morel P, Sanz G et al. International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 1997; 89: 2079–2088.

  2. 2

    Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 1982; 51: 189–199.

  3. 3

    Mufti GJ, Stevens JR, Oscier DG, Hamblin TJ, Machin D . Myelodysplastic syndromes: a scoring system with prognostic significance. Br J Haematol 1985; 59: 425–433.

  4. 4

    Sanz GF, Sanz MA, Vallespi T, Canizo MC, Torrabadella M, Garcia S et al. Two regression models and a scoring system for predicting survival and planning treatment in myelodysplastic syndromes: a multivariate analysis of prognostic factors in 370 patients. Blood 1989; 74: 395–408.

  5. 5

    Morel P, Hebbar M, Lai JL, Duhamel A, Preudhomme C, Wattel E et al. Cytogenetic analysis has strong independent prognostic value in de novo myelodysplastic syndromes and can be incorporated in a new scoring system: a report on 408 cases. Leukemia 1993; 7: 1315–1323.

  6. 6

    Aul C, Gattermann N, Heyll A, Germing U, Derigs G, Schneider W . Primary myelodysplastic syndromes: analysis of prognostic factors in 235 patients and proposals for an improved scoring system. Leukemia 1992; 1: 52–59.

  7. 7

    Toyama K, Ohyashiki K, Yoshida Y, Abe T, Asano S, Hirai H et al. Clinical implications of chromosomal abnormalities in 401 patients with myelodysplastic syndromes: a multicentric study in Japan. Leukemia 1993; 7: 499–508.

  8. 8

    Parlier V, van Melle G, Beris P, Schmidt PM, Tobler A, Haller E et al. Prediction of 18-month survival in patients with primary myelodysplastic syndrome. A regression model and scoring system based on the combination of chromosome findings and the Bournemouth score. Cancer Genet Cytogenet 1995; 81: 158–165.

  9. 9

    Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J et al. World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: report of the Clinical Advisory Committee meeting-Airlie House, Virginia, November 1997. J Clin Oncol 1999; 17: 3835–3849.

  10. 10

    Bennett JM . World Health Organization classification of the acute leukemias and myelodysplastic syndrome. Int J Hematol 2000; 72: 131–133.

  11. 11

    Germing U, Kuendgen A, Gattermann N . Risk assessment in chronic myelomonocytic leukaemia (CMML). Leuk Lymphoma 2004; 34: 1311–1318.

  12. 12

    Pfeilstocker M, Reisner R, Nosslinger T, Gruner H, Nowotny H, Tuchler H et al. Cross-validation of prognostic scores in myelodysplastic syndromes on 386 patients from a single institution confirms importance of cytogenetics. Br J Haematol 1999; 106: 455–463.

  13. 13

    Giagounidis AA, Germing U, Haase S, Hildebrandt B, Schlegelberger B, Schoch C et al. Clinical, morphological, cytogenetic, and prognostic features of patients with myelodysplastic syndromes and del(5q) including band q31. Leukemia 2004; 18: 113–199.

  14. 14

    Aul C, Gattermann N, Germing U, Runde V, Heyll A, Schneider W . Risk assessment in primary myelodysplastic syndromes: validation of the Düsseldorf score. Leukemia 1994; 8: 1906–1913.

  15. 15

    Wimazal F, Sperr WR, Kundi M, Meidlinger P, Fonatsch C, Jordan JH et al. Prognostic value of lactate dehydrogenase activity in myelodysplastic syndromes. Leuk Res 2001; 25: 287–294.

  16. 16

    Sperr WR, Wimazal F, Kundi M, Fonatsch C, Thalhammer-Scherrer R, Schernthaler GH et al. Survival analysis and AML development in patients with de novo myelodysplastic syndromes: comparison of six different prognostic scoring systems. Ann Hematol 2001; 80: 272–275.

  17. 17

    Gonzalez-Medina I, Bueno J, Torrequebrada A, Lopez A, Vallespi T, Massague I . Two groups of chronic myelomonocytic leukaemia: myelodysplastic and myeloproliferative. Prognostic implications in a series of a single center. Leuk Res 2002; 26: 821–824.

  18. 18

    Onida F, Kantarjian HM, Smith TL, Ball G, Keating MJ, Estey EH et al. Prognostic factors and scoring systems in chronic myelomonocytic leukemia: a retrospective analysis of 213 patients. Blood 2002; 99: 840–849.

  19. 19

    Sole F, Espinet B, Sanz GF, Cervera J, Calasanz MJ, Luno E et al. Incidence, characterization and prognostic significance of chromosomal abnormalities in 640 patients with primary myelodysplastic syndromes. Grupo Cooperativo Espanol de Citogenetica Hematologica. Br J Haematol 2000; 108: 346–356.

  20. 20

    Haase D, Fonatsch C, Freund M, Wormann B, Bodenstein H, Bartels H et al. Cytogenetic findings in 179 patients with myelodysplastic syndromes. Ann Hematol 1995; 70: 171–187.

  21. 21

    Oguma S, Yoshida Y, Uchino H, Maekawa T, Nomura T, Mizogushi H . Clinical characteristics of Japanese patients with primary myelodysplastic syndromes: a co-operative study based on 838 cases. Leuk Res 1995; 19: 219–225.

  22. 22

    Lee JH, Lee JH, Shin YR, Lee JS, Kim WK, Chi HS et al. Application of different prognostic scoring systems and comparison of the FAB and WHO classifications in Korean patients with myelodysplastic syndrome. Leukemia 2003; 17: 305–313.

  23. 23

    Germing U, Gattermann N, Strupp C, Aivado M, Aul C . Validation of the WHO proposals for a new classification of primary myelodysplastic syndromes: a retrospective analysis of 1600 patients. Leuk Res 2000; 24: 983–992.

Download references


The authors thank Gisele Rocco, Manuel Aivado, Julie Schanz, Jurij Pitako, Wolfgang R. Sperr, Friedrich Wimazal, and Helga Grüner for their assistance in data collection and karyotyping. This work was supported by Kompetenznetz ‘Akute und chronische Leukämien’, Bundesforschungsministerium (German Research Ministry). R.S. was supported by ‘Tiroler Verein zur Förderung der Krebsforschung an der Universitätsklinik Innsbruck’ und ‘Johannes und Hertha Tuba-Stiftung’, Innsbruck, Austria.

Author information

Correspondence to U Germing.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Germing, U., Hildebrandt, B., Pfeilstöcker, M. et al. Refinement of the international prognostic scoring system (IPSS) by including LDH as an additional prognostic variable to improve risk assessment in patients with primary myelodysplastic syndromes (MDS). Leukemia 19, 2223–2231 (2005) doi:10.1038/sj.leu.2403963

Download citation


  • MDS
  • IPSS
  • LDH
  • WHO-classification
  • FAB-classification

Further reading