Impact of body mass index at different transplantation stages on postoperative outcomes in patients with hematological malignancies: a meta-analysis

Abstract

Although the association between body mass index (BMI) and overall survival (OS) has been reported in leukemia patients of different ages, whether BMI levels at different stages of hematopoietic stem cell transplantation (HSCT) have different effects on postoperative survival remains controversial. We searched four electronic databases from inception through July 2017 without any language restrictions and included studies on different types of hematological malignancies reporting both BMI time points and HSCT. Of the 1420 articles identified, 26 articles were eligible for inclusion in this meta-analysis. Three weight groups (obese, overweight and underweight) were individually compared with the normal group. Summary risk estimates for OS and event-free survival (EFS) were calculated with random- or fixed-effects models. For BMI at the pre-HSCT stage, a statistically significant positive association of increased risk of OS (RR: 1.17; 95% CI: 1.08–1.27) and EFS (RR: 1.29; 95% CI: 1–1.67) was identified in underweight individuals compared with those with normal weights. For BMI in the HSCT stage, a lower BMI was significantly associated with poorer OS (RR: 1.34; 95% CI: 1.01–1.78) and EFS (RR: 1.53; 95% CI: 1.09–2.06) compared with a normal BMI. Our results indicated that lower BMI at the pre-HSCT stage or during HSCT is associated with poorer survival.

Introduction

Although hematopoietic stem cell transplantation (HSCT) is an aggressive therapeutic procedure [1] with a high risk of toxicity and complications, mainly occurring in the gastrointestinal tract [2], it is still the treatment of choice to cure various hematological diseases. Both high-dose chemoradiotherapy and damage to the gastrointestinal tract limit oral intake to a great extent and cause the malabsorption of nutrients, thus compromising a patient’s nutritional status [3]. Conversely, malnourished patients have significantly worse outcomes than well-nourished patients with hematological malignancies [4]. Furthermore, delayed wound healing, a reduced rate of drug metabolism, and impairment of physical and cognitive functions have been reported in malnourished patients [5].

Body mass index (BMI) is frequently used to define nutrition status and explore its association with the risk of mortality [6, 7]. In hematological malignancies, studies have repeatedly revealed that patients with a high BMI have a higher risk of death than those with a low BMI [6, 8, 9]. However, multiple analyses from international consortia have reported inconsistent associations between high BMI and survival [10,11,12,13,14,15]. Recently, some researchers attempted to reconcile these differences via meta-analyses [7,8,9, 16,17,18]. However, these studies ignored the time point for BMI testing when they extracted the data of interest. BMI in different periods is likely to have different effects on the survival of patients with relevant therapeutic operations. Given that finding, in clinical settings, HSCT is usually considered to be the watershed of the entire therapeutic procedure. This raises the question of whether BMI in different transplantation periods may have different effects on overall survival (OS). To reconcile this uncertainty, in this study, we only included articles that were related to both precise BMI time points and HSCT operations. Meanwhile, the time points at which BMI was measured (e.g., pre-HSCT, during HSCT and post-HSCT) were extracted and served as the classification standard.

Patients and methods

Search strategy

Our methodology followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [19]. We systematically searched PubMed, Embase, Ovid Medline and the Cochrane Library to identify studies that investigated BMI and outcomes in patients with hematological malignancies, who were undergoing HSCT, from inception through July 2017. We limited this literature search to articles published in peer-reviewed journals. Our search strategies are included in the Supplemental Table 1.

Eligibility criteria

In this study, we retained studies that reported on subjects diagnosed with hematological malignancies who were treated with an HSCT that reported the effect of weight on treatment outcomes, particularly OS and event-free survival (EFS). If the HSCT did not cover all enrolled patients, the study was excluded from our meta-analysis. Case reports, reviews, editorials, conference abstracts and studies related to animal models or drug therapy were excluded (Fig. 1).

Fig. 1
figure1

Flow of the search strategy

Data extraction

For each included reference, the following information was extracted by two independent investigators: (a) Basic information of the article: country of publication, year, name of first author, research type, quality score and study period; (b) Subject characteristics: gender, age, hematological malignancies, subject number, transplantation type, weight category, median follow-up time, outcomes, adjustments for bias and the time point of BMI data collection; and (c) Statistical data: effect estimates and 95% confidence intervals (CIs).

Weight category

BMI categories were mainly defined according to the WHO standard (“underweight”: BMI <18 kg/m2; “normal”: BMI 18–25 kg/m2; “overweight”: BMI 25–30 kg/m2; “obese”: BMI >30 kg/m2), which accounted for 42.3% (ref. 14, 15, 20,21,22,23,24,25,26,27,28, ref. 28 adopted the WHO standard for Asia) of the included references. Five studies [29,30,31,32,33] used their own weight classification standards based on BMI values (Table 1). The remaining studies used different criteria according to either percentile of BMI [34,35,36,37,38,39], percentage of weight loss [40, 41] or percentage of ideal body weight (BW) [42, 43], and each of these studies is different from the others. We noted that only one study [26] included children <2 years of age, in whom normative values for BMI were not available.

Table 1 Characteristics of studies included in the meta-analysis

Statistical analysis

Summary measures of association (i.e., relative risks, RRs) and 95% CIs were used to evaluate the impact of BMI values that were measured at different time points of HSCT on clinical outcomes. When RRs according to both multivariate and univariate analyses were reported, we only used the value in the multivariate analysis. If measures of association were not presented in the article, RRs and 95% CIs were calculated with the use of reported crude frequencies. Statistical heterogeneity of study effect estimates was calculated using Cochran’s Q test and quantified with the I2statistic [44]. We considered low, moderate and high degrees of heterogeneity to be I2 statistic <25%, 25–50% and ≥50%, respectively. A “leave-one-out” sensitivity analysis was used to evaluate the influence of a particular study on the overall estimates. The quality of literature was assessed using the modified NIH Quality Assessment Tool for Observational Cohort [9] (Supplemental Table 2). Publication bias was assessed using Egger’s tests. All statistical analyses were performed using STATA version 12.0 (State Corporation, College Station, TX, USA). All reported probabilities (p-value) were two sided.

Results

Search descriptions

In total, 1415 articles were identified, and an additional 5 articles were found through Google Scholar and citations of reviews (Fig. 1). After excluding duplicates, 1133 reports were removed by two authors via screening the titles and abstracts. Among the 84 full-text studies that were reviewed, we excluded 12 animal studies, 12 conference abstracts, 8 case reports, 13 reviews, 8 studies that reported that not all subjects underwent HSCT, and an additional 5 references that included patients who were diagnosed with diseases other than hematological malignancies. Finally, 26 studies were identified according to the eligibility criteria and included in this meta-analysis. Assessment of the modified NIH quality score (correction of bias was added) for the included literature revealed that 69.2% of the reports had fair to high quality (≥11/17, modified scoring in reference to NIH scores ≥9/14; Supplemental Table 2). These studies included a total of 32,683 patients with HSCT from 1977 to 2014 (Table 1). Six studies, including a total of 11,720 cases, reported the association between BMI at the time of HSCT and outcomes [14, 20,21,22, 36, 40]. Only one study (145 subjects) investigated the effect of weight loss after HSCT on outcomes [41]. The rest of the studies included 20,818 subjects and evaluated pre-HSCT BMI and outcomes [15, 23,24,25,26,27,28,29,30,31,32,33,34,35, 37,38,39, 42, 43].

Effects of pre-HSCT BMI values on outcomes

Eighteen studies assessed the contribution of pre-HSCT BMI values to outcomes (Table 1). The majority (14) of the studies classified patients into four weight groups (underweight, normal, overweight and obese) based on BMI. Three studies compared the effects of a higher BMI with a lower BMI on outcomes of interest. Only one study reported the association of weight loss with clinical outcomes. Thirteen studies were included in the summary estimate of OS. A statistically significant positive association of increased risk of mortality (OS random-effects RR: 1.17; 95% CI: 1.08–1.27; I2 = 0%) was identified in underweight individuals compared with those of normal weight (Fig. 2). The risk of death decreases with increasing BMI. Interestingly, there were no significant associations of risk of mortality in overweight (OS random-effects RR: 0.99; 95% CI: 0.84–1.17; I2 = 64%) or obese (OS random-effects RR: 0.90; 95% CI: 0.73–1.12; I2 = 70.6%) patients compared with those with normal BMI (Fig. 2). High degrees of heterogeneity were observed for OS in both the overweight and obese groups. The sensitivity analysis, performed by omitting one study at a time, showed that omitting the study by G. Meloni et al. [33] from 11 studies generated a lower heterogeneity (Supplemental Table 3), whereas the omission of the other studies had no substantial influence on the heterogeneity. Visual examination of the funnel plots and an Egger’s test for study effects revealed no evidence of publication bias (OS: under vs. normal p = 0.977; over vs. normal p = 0.238; obese vs. normal p = 0.394).

Fig. 2
figure2

Meta-analyses of overall survival of patients with hematological malignancies according to BMI measured before transplantation. Combined RR was calculated using a random-effects model. The studies are grouped according to different comparison groups (e.g., overweight vs. normal; underweight vs. normal; obese vs. normal), with the black-filled diamonds representing the study-specific RRs and the horizontal line representing the 95% CIs. The open diamonds correspond to summary RRs

Similar results were found in the analyses of the studies reporting EFS (n = 6). There was a tendency for patients with a lower BMI before transplantation to have a poorer EFS (EFS random-effects RR: 1.29; 95% CI: 0.96–1.72; I2 = 17.8%, Fig. 3). No significant between-study heterogeneity was found. A statistically significant association was found when a fixed-effects model was used (EFS fixed-effects RR: 1.29; 95% CI: 1–1.67; I2 = 17.8%). With an increase in BMI, there was a tendency for patients to show improved EFS (EFS random-effects, over vs. normal: 0.91; 95% CI: 0.66–1.25; I2 = 63.8%; EFS random effects, obese vs. normal: 0.74; 95% CI: 0.42–1.27; I2 = 74.4%) to a certain extent (Fig. 3). No obvious publication bias was found using Egger’s test (EFS: under vs. normal p = 0.411; over vs. normal p = 0.325; obese vs. normal p = 0.968).

Fig. 3
figure3

Meta-analyses of event-free survival of patients with hematological malignancies according to BMI measured before transplantation. Combined RR was calculated using a random-effects model. The studies are grouped according to different comparison groups, with the black-filled diamonds representing the study-specific RRs and the horizontal line representing the 95% CIs. The open diamonds correspond to summary RRs

Effects of during HSCT BMI values on outcomes

Six studies evaluated the associations between the levels of BMI at transplantation and outcomes of interest (Table 1). Only one study compared the effect of a higher BMI with a lower BMI on outcomes of interest. The remainder reported on either underweight, overweight or obese compared with normal weight. Similar findings were observed in analyses reporting on underweight patients, defined using BMI values at the time of transplantation, which showed a positive association of BMI with an increased risk of mortality (OS random-effects RR: 1.34; 95% CI: 1.01–1.78; I2 = 60.3%) in those with a lower BMI compared with those with a normal BMI (Fig. 4). Interestingly, a statistically significant association was found in patients with a higher BMI (overweight) at transplantation and a decreased risk of mortality (OS random-effects RR: 0.93; 95% CI: 0.87–0.98; I2 = 12.5%, Fig. 4). No significant association was found for the risk of mortality in obese patients (OS random-effects RR: 0.92; 95% CI: 0.77–1.10; I2 = 73.1%, Fig. 4). Egger’s test for small-study effects revealed no evidence of publication bias (over vs. normal p = 0.298; under vs. normal p = 0.735; obese vs. normal p = 0.882).

Fig. 4
figure4

Meta-analyses of overall survival of patients with hematological malignancies according to BMI measured at the time of transplantation. Combined RR was calculated using a random-effects model. The studies are grouped according to different comparison groups, with the black-filled diamonds representing the study-specific RRs and the horizontal line representing the 95% CIs. The open diamonds correspond to summary RRs

Three studies reported an association between BMI at the time of transplantation and EFS (Table 1). Only two articles provided the data of EFS for patients who were underweight. Pooled estimates were calculated using a random-effects model (EFS random-effects RR: 1.50; 95% CI: 1.09–2.06; I2 = 30.6%, Fig. 5). Lower BMI (underweight) at the time of transplantation was also associated with poorer EFS. Inversely, our data indicated that EFS can be improved in patients with higher BMI values (overweight, EFS random-effects RR: 0.91; 95% CI: 0.85–0.98; I2 = 0%, Fig. 4). However, this association was not observed in obese patients (EFS random-effects RR: 0.85; 95% CI: 0.72–1.01; I2 = 43.4%, Fig. 4). Egger’s test revealed no evidence of publication bias (overweight vs. normal p = 0.412; obese vs. normal p = 0.191).

Fig. 5
figure5

Meta-analyses of event-free survival of patients with hematological malignancies according to BMI measured at the time of transplantation. Combined RR was calculated using a random-effects model. The studies are grouped according to different comparison groups, with the black-filled diamonds representing the study-specific RRs and the horizontal line representing the 95% CIs. The open diamonds correspond to summary RRs

Effects of post-HSCT BMI values on outcomes

To date, the post-HSCT course of BMI and its effect on the outcome of transplantation have not been well studied. Based on our eligibility criteria, we only found one reference that reported an association of post-HSCT BMI and outcomes of interest [41].

Stratification analysis for weight classification

Because the majority (42.3%) of included studies adopted the WHO standard and because each of the remaining studies is different from the others, we used the data from references that adopted the WHO standard to obtain a pooled estimate of the RR. The pre-HSCT stage exhibited a positive association of increased risk of mortality (OS random-effects RR: 1.17; 95% CI: 1.07–1.28; I2 = 0%, Supplemental Fig. 1) in underweight individuals compared with that in normal weight individuals. Only one study was included in the comparison of the EFS of underweight and normal weight individuals. Although no significant associations were found regarding the risk of mortality in overweight or obese patients compared with that in individuals with a normal BMI, the heterogeneities were significantly decreased (overweight vs. normal: OS from 64 to 43.7%, EFS from 63.8 to 0%; obese vs. normal: OS from 70.6 to 0%, EFS from 74.4 to 0%; Supplemental Figs. 1 and 2) when we conducted stratification analyses. During the HSCT period, we observed a positive association of BMI with an increased risk of mortality (OS random-effects RR: 1.43; 95% CI: 1.04–1.97; I2 = 61.9%) in individuals with a lower BMI (underweight) compared with that in individuals with a normal BMI. A statistically significant association was also found for patients with a higher BMI (overweight) at transplantation and a decreased risk of mortality (OS random-effects RR: 0.92; 95% CI: 0.87–0.98; I2 = 12.7%, Supplemental Fig. 3). No significant association was found for the risk of mortality in obese patients (OS random-effects RR: 0.99; 95% CI: 0.87–1.13; I2 = 23.6%, Supplemental Fig. 3). Regarding EFS, the studies used for both the original analysis and the stratification analysis showed consistent results.

Stratification analysis for age

Given that the included references did not clearly distinguish age periods, we divided the studies into a pediatric group (2–17 years of age) and an adult group (>17 years of age) and then performed a meta-regression analysis (Table 2). When the analysis was stratified by age, consistent with the previous findings [45], we observed no statistically significant interaction between the pediatric and adult groups, regardless of transplantation stages (pre-HSCT: p-interaction(under vs. normal) OS = 0.43 and EFS = 0.96; p-interaction(over vs. normal) OS = 0.75 and EFS = 0.08; p-interaction(obese vs. normal) OS = 0.12 and EFS = 0.14; during HSCT: p-interaction(under vs. normal) OS = 0.85 and EFS = NA; p-interaction(over vs. normal) OS = 0.3 and EFS = NA; p-interaction(obese vs. normal) = NA).

Table 2 Stratification analysis for age

Heterogeneity, sensitivity analysis

High degrees of heterogeneity were observed in some subgroups when we compared RRs for OS and EFS. A sensitivity analysis, performed by omitting one study at a time, showed that omitting the study by Meloni et al. from the included studies generated a lower heterogeneity in the pre-HSCT stage (OS: overweight vs. normal I2 = 45.3%; obese vs. normal I2 = 43.4%; EFS: overweight vs. normal I2 = 0%, Supplemental Table 3). Contrarily, omitting Meloni et al. from four studies that evaluated EFS had no substantial influence on the original pooled RRs in the group of obese vs. normal before transplantation (Supplemental Table 3). For during HSCT results, the study by Navarro et al. was a major source of heterogeneity for OS in the group of obese vs. normal patients (new pooled I2 = 0%, Supplemental Table 3). Omitting an unrelated allogeneic cohort in the study by Navarro et al. presented a lower heterogeneity in the group of underweight vs. normal patients (new pooled I2 = 37%, Supplemental Table 3).

Discussion

This meta-analysis included 18 adolescent studies and 6 pediatric studies. Two articles did not distinguish the pediatric patients from adolescents with hematological malignancies [22, 43]. McClune et al. observed that transplantation toxicity, relapse and OS for older adults were not significantly different than those for younger adults undergoing a similar HSCT [45]. When the analysis was stratified by age, we also observed no statistically significant interaction between the pediatric and adult groups, regardless of transplantation stages (Table 2). Therefore, age was not included in the inclusion criteria in this transplantation-related meta-analysis. Additionally, it was impossible to analyze the differences between the four weight groups for the outcomes of interest because the identified studies did not provide sufficient data. To reconcile this contradiction, we only compared the effect of abnormal BMI (underweight, overweight and obese) with normal BMI on outcomes of interest.

Before transplantation (pre-HSCT), we observed worse OS and EFS in patients with a lower BMI than in those with a normal BMI (Figs. 2 and 3). With the increase in BMI, the risk of mortality after transplantation showed a decreasing trend (Figs. 2 and 3). There are several possible explanations. Normally, patients with a lower BMI exhibit a lower tolerance to a series of therapeutic treatments, such as radiotherapy and myeloablative chemotherapy, before HSCT, thus resulting in a higher rate of mortality. Furthermore, lower BMI may reflect the presence of more aggressive disease, causing a greater degree of physiological derangement. It is also possible that patient selection bias may influence these findings. Data are not available for patients who did not undergo transplantation. Similar findings were observed in the analyses of OS and EFS, with a positive association of an increased risk of mortality in those with a lower BMI measured at the time of transplantation compared with those with a normal BMI. There is generally a survival disadvantage for both overweight and underweight patients with hematological malignancies. However, our data contradict this conventional viewpoint. In this study, better OS was experienced by the overweight or obese patients compared with the underweight patients. When taking transplantation into consideration, patients with a higher BMI may be likely to have more advantages in tolerance to transplantation-related therapy. Notably, the introduction of reduced intensity conditioning (RIC) has expanded the recipient pool for transplantation, which has made transplant an option for older individuals and those with more severe conditions and is thought to decrease the risk of early morbidity [46]. The effect of treatment-related factors on mortality was greater than that of malnutrition-related factors. However, malnutrition can exacerbate the risk of death under the same conditions [6, 8, 9, 47, 48]. In this study, both myeloablative conditioning (MAC) and RIC regimens can be found in the included references. Given that none of the studies have a specific distinction between MAC and RIC, we cannot perform a meta-regression analysis to evaluate the effects of these two regimens on the overall outcomes.

To address whether BMI in different transplantation periods may have different effects on OS, we only considered two main factors for transplantation and BMI time points to perform this meta-analysis. Correspondingly, there were several limitations. First, it is difficult to normalize the weight classification standard because the identified studies did not provide detailed information about individual BMI values. Therefore, we used their own weight classifications to conduct the analysis. To evaluate whether different weight classification standards affected the combined estimate results, we performed a subgroup analysis. The results of the subgroup analysis (Supplemental Figs. 1, 2 and 3) of studies adopting the WHO standard are consistent with previous analyses using combined weight classification standards (Figs. 2, 3 and 4). Thus, two conclusions can be drawn: (1) In this study, different weight classification criteria did not have a significant effect on the pooled RR estimate. (2) Our findings apply to an HSCT cohort using the WHO weight classification standard. Second, the inclusion and pooling of such varied hematological malignancies conditions are likely to introduce heterogeneity. Initially, we attempted to incorporate limitations of the type of hematological malignancy into our search strategy. However, the references were not sufficient to conduct this meta-analysis, which did not allow us to focus on a specific type of hematological malignancy. Additionally, we also attempted to segregate references into several hematological malignancy subgroups according to the WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues” by Swerdlow, et al. Similarly, we could not obtain specific data on each identified disease.

In summary, our study indicated that patients who are underweight at the pre-HSCT stage or during HSCT have significantly poorer outcomes. With the increase in related studies involving a single type of hematological malignancy, a comprehensive meta-analysis focused on BMIs at different stages of HSCT is warranted.

References

  1. 1.

    Gomez Alvarez ME. [Parenteral nutrition in hematopoietic stem cell transplantation]. Farm Hosp. 2004;28:116–22.

  2. 2.

    Martin-Salces M, de Paz R, Canales MA, Mesejo A, Hernandez-Navarro F. Nutritional recommendations in hematopoietic stem cell transplantation. Nutrition. 2008;24:769–75. https://doi.org/10.1016/j.nut.2008.02.021.

  3. 3.

    Garcia-Luna PP, Parejo Campos J, Pereira Cunill JL. [Causes and impact of hyponutrition and cachexia in the oncologic patient]. Nutr Hosp. 2006;21(Suppl 3):10–16.

  4. 4.

    Lobato-Mendizabal E, Ruiz-Arguelles GJ, Marin-Lopez A. Leukaemia and nutrition. I: malnutrition is an adverse prognostic factor in the outcome of treatment of patients with standard-risk acute lymphoblastic leukaemia. Leuk Res. 1989;13:899–906.

  5. 5.

    Sullivan DH. The role of nutrition in increased morbidity and mortality. Clin Geriatr Med. 1995;11:661–74.

  6. 6.

    Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. 2003;348:1625–38. https://doi.org/10.1056/NEJMoa021423.

  7. 7.

    Larsson SC, Wolk A. Overweight and obesity and incidence of leukemia: a meta-analysis of cohort studies. Int J Cancer. 2008;122:1418–21. https://doi.org/10.1002/ijc.23176.

  8. 8.

    Castillo JJ, Reagan JL, Ingham RR, Furman M, Dalia S, Merhi B, et al. Obesity but not overweight increases the incidence and mortality of leukemia in adults: a meta-analysis of prospective cohort studies. Leuk Res. 2012;36:868–75. https://doi.org/10.1016/j.leukres.2011.12.020.

  9. 9.

    Orgel E, Genkinger JM, Aggarwal D, Sung L, Nieder M, Ladas EJ. Association of body mass index and survival in pediatric leukemia: a meta-analysis. Am J Clin Nutr. 2016;103:808–17. https://doi.org/10.3945/ajcn.115.124586.

  10. 10.

    Gelelete CB, Pereira SH, Azevedo AM, Thiago LS, Mundim M, Land MG, et al. Overweight as a prognostic factor in children with acute lymphoblastic leukemia. Obes (Silver Spring). 2011;19:1908–11. https://doi.org/10.1038/oby.2011.195.

  11. 11.

    Hijiya N, Panetta JC, Zhou Y, Kyzer EP, Howard SC, Jeha S, et al. Body mass index does not influence pharmacokinetics or outcome of treatment in children with acute lymphoblastic leukemia. Blood. 2006;108:3997–4002. https://doi.org/10.1182/blood-2006-05-024414.

  12. 12.

    Aldhafiri FK, McColl JH, Reilly JJ. Prognostic significance of being overweight and obese at diagnosis in children with acute lymphoblastic leukemia. J Pediatr Hematol Oncol. 2014;36:234–6. https://doi.org/10.1097/MPH.0000000000000056.

  13. 13.

    Baillargeon J, Langevin AM, Lewis M, Estrada J, Mullins J, Pitney A, et al. Obesity and survival in a cohort of predominantly Hispanic children with acute lymphoblastic leukemia. J Pediatr Hematol Oncol. 2006;28:575–8. https://doi.org/10.1097/01.mph.0000212985.33941.d8.

  14. 14.

    Nikolousis E, Nagra S, Paneesha S, Delgado J, Holder K, Bratby L, et al. Allogeneic transplant outcomes are not affected by body mass index (BMI) in patients with haematological malignancies. Ann Hematol. 2010;89:1141–5. https://doi.org/10.1007/s00277-010-1001-6.

  15. 15.

    Jaime-Perez JC, Colunga-Pedraza PR, Gutierrez-Gurrola B, Brito-Ramirez AS, Gutierrez-Aguirre H, Cantu-Rodriguez OG, et al. Obesity is associated with higher overall survival in patients undergoing an outpatient reduced-intensity conditioning hematopoietic stem cell transplant. Blood Cells Mol Dis. 2013;51:61–65.

  16. 16.

    Amankwah EK, Saenz AM, Hale GA, Brown PA. Association between body mass index at diagnosis and pediatric leukemia mortality and relapse: a systematic review and meta-analysis. Leuk Lymphoma. 2016;57:1140–8. https://doi.org/10.3109/10428194.2015.1076815.

  17. 17.

    Nakao M, Chihara D, Niimi A, Ueda R, Tanaka H, Morishima Y, et al. Impact of being overweight on outcomes of hematopoietic SCT: a meta-analysis. Bone Marrow Transplant. 2014;49:66–72. https://doi.org/10.1038/bmt.2013.128.

  18. 18.

    Larsson SC, Wolk A. Obesity and risk of non-Hodgkin’s lymphoma: a meta-analysis. Int J Cancer. 2007;121:1564–70. https://doi.org/10.1002/ijc.22762.

  19. 19.

    Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1 https://doi.org/10.1186/2046-4053-4-1.

  20. 20.

    Navarro WH, Loberiza FR Jr., Bajorunaite R, van Besien K, Vose JM, Lazarus HM, et al. Effect of body mass index on mortality of patients with lymphoma undergoing autologous hematopoietic cell transplantation. Biol Blood Marrow Transplant: J Am Soc Blood Marrow Transplant. 2006;12:541–51.

  21. 21.

    Navarro WH, Agovi MA, Logan BR, Ballen K, Bolwell BJ, Frangoul H, et al. Obesity does not preclude safe and effective myeloablative hematopoietic cell transplantation (HCT) for acute myelogenous leukemia (AML) in adults. Biol Blood Marrow Transplant. 2010;16:1442–50. https://doi.org/10.1016/j.bbmt.2010.04.009.

  22. 22.

    Vogl DT, Wang T, Pérez WS, Stadtmauer EA, Heitjan DF, Lazarus HM, et al. Effect of obesity on outcomes after autologous hematopoietic stem cell transplantation for multiple myeloma. Biol Blood Marrow Transplant. 2011;17:1765–74. https://doi.org/10.1016/j.bbmt.2011.05.005.

  23. 23.

    Lee HJ, Licht AS, Hyland AJ, Ford LA, Sait SNJ, Block AW, et al. Is obesity a prognostic factor for acute myeloid leukemia outcome? Ann Hematol. 2012;91:359–65. https://doi.org/10.1007/s00277-011-1319-8.

  24. 24.

    Sucak GT, Suyan IE, Baysal NA, AltIndal Ş, Çakar MK, Ak IŞZ, et al. The role of body mass index and other body composition parameters in early post-transplant complications in patients undergoing allogeneic stem cell transplantation with busulfan-cyclophosphamide conditioning. Int J Hematol. 2012;95:95–101. https://doi.org/10.1007/s12185-011-0980-y.

  25. 25.

    Fuji S, Takano K, Mori T, Eto T, Taniguchi S, Ohashi K, et al. Impact of pretransplant body mass index on the clinical outcome after allogeneic hematopoietic SCT. Bone Marrow Transplant. 2014;49:1505–12. https://doi.org/10.1038/bmt.2014.178.

  26. 26.

    Gleimer M, Li Y, Chang L, Paczesny S, Hanauer DA, Frame DG, et al. Baseline body mass index among children and adults undergoing allogeneic hematopoietic cell transplantation: clinical characteristics and outcomes. Bone Marrow Transplant. 2015;50:402–10. https://doi.org/10.1038/bmt.2014.280.

  27. 27.

    Lau JE, Weber C, Earl M, Rybicki LA, Carlstrom KD, Wenzell CM, et al. Outcomes after autologous SCT in lymphoma patients grouped by weight. Bone Marrow Transplant. 2015;50:652–7. https://doi.org/10.1038/bmt.2014.327.

  28. 28.

    Yang J, Xue SL, Zhang X, Zhou YN, Qin LQ, Shen YP, et al. Effect of body mass index on overall survival of patients with allogeneic hematopoietic stem cell transplantation. Eur J Clin Nutr. 2017;71:750–4. https://doi.org/10.1038/ejcn.2016.225.

  29. 29.

    Le Blanc K, Ringden O, Remberger M. A low body mass index is correlated with poor survival after allogeneic stem cell transplantation. Haematologica. 2003;88:1044–52.

  30. 30.

    Baumgartner A, Zueger N, Bargetzi A, Medinger M, Passweg JR, Stanga Z, et al. Association of nutritional parameters with clinical outcomes in patients with acute myeloid leukemia undergoing haematopoietic stem cell transplantation. Ann Nutr Metab. 2016;69:89–98. https://doi.org/10.1159/000449451.

  31. 31.

    Hadjibabaie M, Tabeefar H, Alimoghaddam K, Iravani M, Eslami K, Honarmand H, et al. The relationship between body mass index and outcomes in leukemic patients undergoing allogeneic hematopoietic stem cell transplantation. Clin Transplant. 2012;26:149–55. https://doi.org/10.1111/j.1399-0012.2011.01445.x.

  32. 32.

    Dietrich S, Radujkovic A, Stolzel F, Falk CS, Benner A, Schaich M, et al. Pretransplant metabolic distress predicts relapse of acute myeloid leukemia after allogeneic stem cell transplantation. Transplantation. 2015;99:1065–71. https://doi.org/10.1097/TP.0000000000000471.

  33. 33.

    Meloni G, Proia A, Capria S, Romano A, Trape G, Trisolini SM, et al. Obesity and autologous stem cell transplantation in acute myeloid leukemia. Bone Marrow Transplant. 2001;28:365–7.

  34. 34.

    Urbain P, Birlinger J, Ihorst G, Biesalski H-K, Finke J, Bertz H. Body mass index and bioelectrical impedance phase angle as potentially modifiable nutritional markers are independent risk factors for outcome in allogeneic hematopoietic cell transplantation. Ann Hematol. 2013;92:111–9.

  35. 35.

    Bulley S, Gassas A, Dupuis LL, Aplenc R, Beyene J, Greenberg ML, et al. Inferior outcomes for overweight children undergoing allogeneic stem cell transplantation. Br J Haematol. 2008;140:214–7.

  36. 36.

    Barker CC, Agovi MA, Logan B, Lazarus HM, Ballen KK, Gupta V, et al. Childhood obesity and outcomes after bone marrow transplantation for patients with severe aplastic anemia. Biol Blood Marrow Transplant: J Am Soc Blood Marrow Transplant. 2011;17:737–44. https://doi.org/10.1016/j.bbmt.2010.08.019. e-pub ahead of print 2010/09/08.

  37. 37.

    Aplenc R, Zhang MJ, Sung L, Zhu X, Ho VT, Cooke K, et al. Effect of body mass in children with hematologic malignancies undergoing allogeneic bone marrow transplantation. Blood. 2014;123:3504–11. https://doi.org/10.1182/blood-2013-03-490334.

  38. 38.

    Pine M, Wang L, Harrell FE Jr., Calder C, Manes B, Evans M, et al. The effect of obesity on outcome of unrelated cord blood transplant in children with malignant diseases. Bone Marrow Transplant. 2011;46:1309–13.

  39. 39.

    Hoffmeister PA, Storer BE, Macris PC, Carpenter PA, Baker KS. Relationship of body mass index and arm anthropometry to outcomes after pediatric allogeneic hematopoietic cell transplantation for hematologic malignancies. Biol Blood Marrow Transplant. 2013;19:1081–6. https://doi.org/10.1016/j.bbmt.2013.04.017.

  40. 40.

    Radujkovic A, Becker N, Benner A, Penack O, Platzbecker U, Stolzel F, et al. Pre-transplant weight loss predicts inferior outcome after allogeneic stem cell transplantation in patients with myelodysplastic syndrome. Oncotarget. 2015;6:35095–106. https://doi.org/10.18632/oncotarget.4805. e-pub ahead of print 2015/09/12.

  41. 41.

    Fuji S, Mori T, Khattry N, Cheng J, Do YR, Yakushijin K, et al. Severe weight loss in 3 months after allogeneic hematopoietic SCT was associated with an increased risk of subsequent non-relapse mortality. Bone Marrow Transplant. 2015;50:100–5. https://doi.org/10.1038/bmt.2014.228.

  42. 42.

    White M, Murphy AJ, Hallahan A, Ware RS, Fraser C, Davies PSW. Survival in overweight and underweight children undergoing hematopoietic stem cell transplantation. Eur J Clin Nutr. 2012;66:1120–3. https://doi.org/10.1038/ejcn.2012.109.

  43. 43.

    Fleming DR, Rayens MK, Garrison J. Impact of obesity on allogeneic stem cell transplant patients: a matched case-controlled study. Am J Med. 1997;102:265–8. https://doi.org/10.1016/S0002-9343(96)00450-0.

  44. 44.

    Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–60. https://doi.org/10.1136/bmj.327.7414.557.

  45. 45.

    McClune BL, Weisdorf DJ, Pedersen TL, Tunes da Silva G, Tallman MS, Sierra J, et al. Effect of age on outcome of reduced-intensity hematopoietic cell transplantation for older patients with acute myeloid leukemia in first complete remission or with myelodysplastic syndrome. J Clin Oncol. 2010;28:1878–87. https://doi.org/10.1200/JCO.2009.25.4821.

  46. 46.

    Sengsayadeth S, Savani BN, Blaise D, Malard F, Nagler A, Mohty M. Reduced intensity conditioning allogeneic hematopoietic cell transplantation for adult acute myeloid leukemia in complete remission—a review from the Acute Leukemia Working Party of the EBMT. Haematologica. 2015;100:859–69. https://doi.org/10.3324/haematol.2015.123331.

  47. 47.

    Prospective Studies C, Whitlock G, Lewington S, Sherliker P, Clarke R, Emberson J, et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet. 2009;373:1083–96. https://doi.org/10.1016/S0140-6736(09)60318-4.

  48. 48.

    He J, McGee D, Niu X, Choi W. Examining the dynamic association of BMI and mortality in the Framingham Heart Study. Int J Environ Res Public Health. 2009;6:3115–26. https://doi.org/10.3390/ijerph6123115.

Download references

Acknowledgements

We thank the anonymous reviewers for their helpful suggestions. This work was supported by the National Natural Science Foundation of China (no. 81703221 to GR) and the Special Fund for Basic Scientific Research (no. 1610422016003 to GR and no. 1610422017007 to GR). ORCID profile (Dr. Guangxu Ren): 0000-0003-2378-260×.

Author contributions

GR designed the research. GR and WC conducted the research. SY, LW, LL, JH and JW analyzed the data. GR drafted the manuscript.

Author information

Correspondence to Guangxu Ren.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Electronic supplementary material

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ren, G., Cai, W., Wang, L. et al. Impact of body mass index at different transplantation stages on postoperative outcomes in patients with hematological malignancies: a meta-analysis. Bone Marrow Transplant 53, 708–721 (2018). https://doi.org/10.1038/s41409-018-0234-1

Download citation

Further reading